Author: "Department of Data Analysis" - Searchworks@Jio Institute Digital Library Search Results

Your search keyword '"Department of Data Analysis"' showing total 672 results

Start Over Author "Department of Data Analysis"

672 results on '"Department of Data Analysis"'

1. Axiomatic characterization of the χ 2 dissimilarity measure

Author: Bouyssou, Denis, Marchant, Thierry, Pirlot, Marc, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Université de Mons (UMons), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
Subjects: Computer Science::Logic in Computer Science, [SHS.ECO]Humanities and Social Sciences/Economics and Finance
Abstract: We axiomatically characterize the χ 2 dissimilarity measure. To this end, we solve a new generalization of a functional equation discussed in Aczel (Lectures on functional equations and their applications, Academic Press, 1966).
Published: 2021

2. A theoretical look at ELECTRE TRI-nB

Author: Bouyssou, Denis, Marchant, Thierry, Pirlot, Marc, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Universiteit Gent = Ghent University [Belgium] (UGENT), Faculté polytechnique de Mons, Université de Mons (UMons), Bouyssou, Denis, Department of Data Analysis, Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
Subjects: [SHS.PSY] Humanities and Social Sciences/Psychology, Multiple criteria analysis, [INFO.INFO-RO] Computer Science [cs]/Operations Research [cs.RO], ELECTRE TRI-B, Sorting models, [SHS.PSY]Humanities and Social Sciences/Psychology, ELECTRE TRI-nB, ELECTRE TRI, [INFO.INFO-RO]Computer Science [cs]/Operations Research [cs.RO], [SHS.ECO] Humanities and Social Sciences/Economics and Finance, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, Decision with multiple attributes
Abstract: ELECTRE TRI is a set of methods designed to sort alternatives evaluated on several attributes into ordered categories. The original ELEC-TRE TRI-B method uses one limiting profile per category. A more recent method, ELECTRE TRI-nB, allows one to use several limiting profiles for each category. We investigate the properties of ELECTRE TRI-nB. When the number of limiting profiles used to define each category is not restricted, ELECTRE TRI-nB is easy to characterize axiomatically and is found to be equivalent to several other methods proposed in the literature. We extend this result in various directions., Ce texte propose une analyse axiomatique de la méthode ELECTRE TRI-nB visant à affecter des objets évalués sur plusieurs critères dans des catégories ordonnées. Cette analyse axiomatique permet de la positionner vis-àvis d'autre méthodes du même type.
Published: 2020

3. A theoretical look at ELECTRE TRI-nB a

Author: Bouyssou, Denis, Marchant, Thierry, Pirlot, Marc, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Ghent University [Belgium] (UGENT), Faculté polytechnique de Mons, and Université de Mons (UMons)
Subjects: ELECTRE TRI-B, Sorting models, ELECTRE TRI-nB, [SHS.PSY]Humanities and Social Sciences/Psychology, [INFO.INFO-RO]Computer Science [cs]/Operations Research [cs.RO], [SHS.ECO]Humanities and Social Sciences/Economics and Finance, Decision with multiple attributes
Abstract: ELECTRE TRI is a set of methods designed to sort alternatives evaluated on several attributes into ordered categories. The original ELEC-TRE TRI-B method uses one limiting profile per category. A more recent method, ELECTRE TRI-nB, allows one to use several limiting profiles for each category. We investigate the properties of ELECTRE TRI-nB. When the number of limiting profiles used to define each category is not restricted, ELECTRE TRI-nB is easy to characterize axiomatically and is found to be equivalent to several other methods proposed in the literature. We extend this result in various directions.; Ce texte propose une analyse axiomatique de la méthode ELECTRE TRI-nB visant à affecter des objets évalués sur plusieurs critères dans des catégories ordonnées. Cette analyse axiomatique permet de la positionner vis-àvis d'autre méthodes du même type.
Published: 2020

4. Data Makes the Public Sector Go Round

Author: Sofia Sgagia, Aikaterini-Vasiliki Douka, Stefan Decker, Charalampia Komopoulou, Dimosthenis Vardouniotis, Vaios Tsarapatsanis, Paraskevi Christodoulou, Vassilios Peristeras, International Hellenic University, Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Peter Parycek, Olivier Glassey, Marijn Janssen, Hans Jochen Scholl, Efthimios Tambouris, Evangelos Kalampokis, Shefali Virkar, TC 8, and WG 8.5
Subjects: Data, Public Sector, Public Administration, business.industry, Data management, [SHS.INFO]Humanities and Social Sciences/Library and information sciences, 05 social sciences, Public sector, Usability, 02 engineering and technology, Public relations, Transparency (behavior), 0506 political science, Politics, Systematic review, 020204 information systems, Public participation, 050602 political science & public administration, 0202 electrical engineering, electronic engineering, information engineering, [INFO]Computer Science [cs], Business, Literature Review
Abstract: Part 2: Open Data, Linked Data, and Semantic Web; International audience; This article summarizes the results of a systematic literature review on how the new data technologies affect the public sector and what their impact on governments are. The opportunities and the challenges that public administrations face nowadays in a data-driven world are important. Our research retrieved opportunities related to transparency, innovation, public participation and efficiency, while challenges regarding privacy issues, technical difficulties, data management, cultural and political obstacles were found. By overcoming the challenges and empowering the opportunities for data usability in public sector, exploitation of data could become the foundation for innovation and public sector transformation.
Published: 2018
Full Text: View/download PDF

5. The β -ranking and the β -measure for directed networks: Axiomatic characterizations

Author: Denis Bouyssou, Thierry Marchant, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL), Ghent University (UGent), Department of Analytical Chemistry, Université Paris Dauphine-PSL-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Ghent University [Belgium] (UGENT), Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS)
Subjects: Sociology and Political Science, business.industry, 05 social sciences, Directed network, General Social Sciences, Characterization (mathematics), 050905 science studies, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, Measure (mathematics), Ranking (information retrieval), Anthropology, [INFO]Computer Science [cs], Artificial intelligence, 0509 other social sciences, Relational power measure, Beta-measure, 050904 information & library sciences, business, General Psychology, Axiom, ComputingMilieux_MISCELLANEOUS, Mathematics
Abstract: International audience; The β-measure has been introduced and characterized by van den Brink and Gilles (2000) for measuring the domination in directed networks. The present paper characterizes the ranking induced by the β-measure. It also provides an alternative characterization of the β-measure.
Published: 2018
Full Text: View/download PDF

6. Conditional Expected Utility Criteria for Decision Making under Ignorance or Objective Ambiguity

Author: Nicolas Gravel, Arunava Sen, Thierry Marchant, Groupement de Recherche en Économie Quantitative d'Aix-Marseille (GREQAM), École Centrale de Marseille (ECM)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Aix Marseille Université (AMU), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Indian Statistical Institute [New Delhi], Lai Tong, Charles, Aix-Marseille Sciences Economiques (AMSE), École des hautes études en sciences sociales (EHESS)-Aix Marseille Université (AMU)-École Centrale de Marseille (ECM)-Centre National de la Recherche Scientifique (CNRS), Ghent University [Belgium] (UGENT), École des hautes études en sciences sociales (EHESS)-École Centrale de Marseille (ECM)-Centre National de la Recherche Scientifique (CNRS)-Aix Marseille Université (AMU), and Universiteit Gent = Ghent University (UGENT)
Subjects: Economics and Econometrics, Computer science, media_common.quotation_subject, expected utility, ranking sets, Set (abstract data type), 0502 economics and business, [MATH]Mathematics [math], 050207 economics, [SHS.ECO] Humanities and Social Sciences/Economics and Finance, Axiom, Expected utility hypothesis, 050205 econometrics, media_common, Applied Mathematics, axioms, 05 social sciences, Conditional probability, conditional probabilities, Ambiguity, Function (mathematics), 16. Peace & justice, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, D81, ignorance, axioms JEL classification numbers: D80, Ranking, ambiguity, Probability distribution, Mathematical economics
Abstract: Corresponding publicationsWorking Paper | Conditional Expected Utility Criteria for Decision Making under Ignorance or Objective Ambiguity ; International audience; We provide an axiomatic characterization of a family of criteria for ranking completely uncertain and/or ambiguous decisions. A completely uncertain decision is described by the set of all its consequences (assumed to be finite). An ambiguous decision is described as a set of possible probability distributions over a set of prizes. Every criterion in the family compares sets on the basis of their conditional expected utility , for some “likelihood” function taking strictly positive values and some utility function both having the universe of alternatives as their domain.
Published: 2016
Full Text: View/download PDF

7. Ranking authors using fractional counting of citations: An axiomatic approach

Author: Denis Bouyssou, Thierry Marchant, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Ghent University [Belgium] (UGENT), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Universiteit Gent = Ghent University [Belgium] (UGENT), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
Subjects: Normalization (statistics), Fractional counting, Normalisation, Library and Information Sciences, 050905 science studies, Computer Science::Digital Libraries, [INFO]Computer Science [cs], Fractional counting of citations, GeneralLiterature_REFERENCE(e.g.,dictionaries,encyclopedias,glossaries), Axiom, ComputingMilieux_MISCELLANEOUS, Mathematics, Information retrieval, 05 social sciences, Bibliometric ranking of authors, Axiomatic system, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, Computer Science Applications, Mathematics and Statistics, Axiomatization, Rangement, Bibliométrie, 0509 other social sciences, 050904 information & library sciences, Citation, Field normalization
Abstract: International audience; This paper analyzes from an axiomatic point of view a recent proposal for counting citations: the value of a citation given by a paper is inversely proportional to the total number of papers it cites. This way of fractionally counting citations was suggested as a possible way to normalize citation counts between fields of research having different citation cultures. It belongs to the “citing-side” approach to normalization. We focus on the properties characterizing this way of counting citations when it comes to ranking authors. Our analysis is conducted within a formal framework that is more complex but also more realistic than the one usually adopted in most axiomatic analyses of this kind.
Published: 2016
Full Text: View/download PDF

8. How Wrong Can We Get? A Review of Machine Learning Approaches and Error Bars

Author: Gilles Blanchard, Timon Schroeter, Anton Schwaighofer, Sebastian Mika, Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), and Publica
Subjects: Computer science, media_common.quotation_subject, Normal Distribution, Decision tree, Quantitative Structure-Activity Relationship, Ligands, 010402 general chemistry, computer.software_genre, Machine learning, 01 natural sciences, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Relevance vector machine, 03 medical and health sciences, symbols.namesake, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Artificial Intelligence, [MATH.MATH-ST]Mathematics [math]/Statistics [math.ST], Drug Discovery, Computer Simulation, Quality (business), Gaussian process, ComputingMilieux_MISCELLANEOUS, 030304 developmental biology, media_common, 0303 health sciences, Models, Statistical, business.industry, Model selection, Decision Trees, Organic Chemistry, General Medicine, 0104 chemical sciences, Computer Science Applications, Support vector machine, Models, Chemical, symbols, Data mining, Artificial intelligence, Focus (optics), business, computer, Model building, Algorithms
Abstract: A large number of different machine learning methods can potentially be used for ligand-based virtual screening. In our contribution, we focus on three specific nonlinear methods, namely support vector regression, Gaussian process models, and decision trees. For each of these methods, we provide a short and intuitive introduction. In particular, we will also discuss how confidence estimates (error bars) can be obtained from these methods. We continue with important aspects for model building and evaluation, such as methodologies for model selection, evaluation, performance criteria, and how the quality of error bar estimates can be verified. Besides an introduction to the respective methods, we will also point to available implementations, and discuss important issues for the practical application.
Published: 2009
Full Text: View/download PDF

9. Finite-Dimensional Projection for Classification and Statistical Learning

Author: Gilles Blanchard, Laurent Zwald, Publica, Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Learning and recognition in vision (LEAR), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Jean Kuntzmann (LJK), Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Centre National de la Recherche Scientifique (CNRS)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Centre National de la Recherche Scientifique (CNRS), Statistique et Modélisation Stochatisque (SMS), and Laboratoire Jean Kuntzmann (LJK)
Subjects: dimension reduction, 02 engineering and technology, Library and Information Sciences, 01 natural sciences, Regularization (mathematics), 010104 statistics & probability, [MATH.MATH-ST]Mathematics [math]/Statistics [math.ST], 0202 electrical engineering, electronic engineering, information engineering, 0101 mathematics, Mathematics, business.industry, Dimensionality reduction, 020206 networking & telecommunications, Pattern recognition, [STAT.TH]Statistics [stat]/Statistics Theory [stat.TH], kernel principal component analysis (KPCA), Linear subspace, Computer Science Applications, regularization, Support vector machine, statistical learning, Statistical classification, Kernel method, classification, Rate of convergence, Binary classification, support vector machine (SVM), Artificial intelligence, business, Algorithm, Information Systems
Abstract: In this paper, a new method for the binary classification problem is studied. It relies on empirical minimization of the hinge risk over an increasing sequence of finite-dimensional spaces. A suitable dimension is picked by minimizing the regularized risk, where the regularization term is proportional to the dimension. An oracle-type inequality is established for the excess generalization risk (i.e., regret to Bayes) of the procedure, which ensures adequate convergence properties of the method. We suggest to select the considered sequence of subspaces by applying kernel principal components analysis (KPCA). In this case, the asymptotical convergence rate of the method can be better than what is known for the support vector machine (SVM). Exemplary experiments are presented on benchmark data sets where the practical results of the method are comparable to the SVM.
Published: 2008
Full Text: View/download PDF

10. An axiomatic approach to noncompensatory sorting methods in MCDM, I: The case of two categories

Author: Thierry Marchant, Denis Bouyssou, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
Subjects: 021103 operations research, Information Systems and Management, Sorting algorithm, General Computer Science, 0211 other engineering and technologies, Axiomatic system, 02 engineering and technology, Management Science and Operations Research, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, Multiple-criteria decision analysis, computer.software_genre, Industrial and Manufacturing Engineering, Axiomatic analysis, Modeling and Simulation, 0202 electrical engineering, electronic engineering, information engineering, sort, Multicriteria analysis, 020201 artificial intelligence & image processing, Data mining, ELECTRE, computer, ComputingMilieux_MISCELLANEOUS, Axiom, Mathematics
Abstract: In the literature on MCDM, many methods have been proposed in order to sort alternatives evaluated on several attributes into ordered categories. Most of them were proposed on an ad hoc basis. The purpose of this paper is to contribute to a recent trend of research aiming at giving these methods sound theoretical foundations. Using tools from conjoint measurement, we provide an axiomatic analysis of the partitions of alternatives into two categories that can be obtained using what we call “noncompensatory sorting models”. These models have strong links with the pessimistic version of ELECTRE TRI. Our analysis allows to pinpoint what appears to be the main distinctive features of ELECTRE TRI when compared to other sorting methods. It also gives hints on the various methods that have been proposed to assess the parameters of ELECTRE TRI on the basis of assignment examples.
Published: 2007
Full Text: View/download PDF

11. Modelling preferences

Author: Denis Bouyssou, Thierry Marchant, Marc Pirlot, Alexis Tsoukiàs, Philippe Vincke, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Faculté Polytechnique (UMONS), Université de Mons (UMons), Université Libre de Bruxelles, and Université libre de Bruxelles (ULB)
Subjects: preference modelling, 021103 operations research, 05 social sciences, 0211 other engineering and technologies, 0501 psychology and cognitive sciences, [INFO]Computer Science [cs], 02 engineering and technology, decision aiding, 050105 experimental psychology
Abstract: International audience; This chapter deals with a crucial step in the decision aiding process: the aggregation of the alternatives’ performances on each criterion in order to faithfully model the overall preference of the decision maker. The approach we follow is that of conjoint measurement, which aims at determining under which conditions a preference can be represented in a particular aggregation model. This approach is first illustrated with the classical additive value function model. Then, we describe two broad families of preference models, which constitute a framework encompassing many aggregation models used in practice. The aggregation rules that fit with the second family of models rely on the aggregation of preference differences. Among this family we find, in particular, models for the outranking relations (concordance relations with vetoes) that are used in several case studies in this book.
Published: 2015
Full Text: View/download PDF

12. Bi-semiorders with frontiers on finite sets

Author: Denis Bouyssou, Thierry Marchant, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, and Universiteit Gent = Ghent University [Belgium] (UGENT)
Subjects: Conjoint measurement, Applied Mathematics, Dimension (graph theory), Semiorder, Interval order, Extension (predicate logic), 16. Peace & justice, Bi-semiorder, Frontier, Combinatorics, Product (mathematics), Biorder, [INFO]Computer Science [cs], Finite set, General Psychology, Mathematics
Abstract: Document de travail: Cahier du LAMSADE, n°348 (2013); International audience; This paper studies an extension of bi-semiorders in which a “frontier” is added between the various relations used. This extension is motivated by the study of additive representations of ordered partitions and coverings defined on product sets of two components.
Published: 2015
Full Text: View/download PDF

13. Statistical properties of kernel principal component analysis

Author: Laurent Zwald, Gilles Blanchard, Olivier Bousquet, Publica, Département de Mathématiques [ORSAY], Université Paris-Sud - Paris 11 (UP11), Max Planck Institute for Biological Cybernetics, Max-Planck-Gesellschaft, Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Département de Mathématiques-Université de Paris XI, Learning and recognition in vision (LEAR), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Jean Kuntzmann (LJK), Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Centre National de la Recherche Scientifique (CNRS)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Centre National de la Recherche Scientifique (CNRS), Statistique et Modélisation Stochatisque (SMS), Laboratoire Jean Kuntzmann (LJK), and Zwald, Laurent
Subjects: Kernel principal components analysis, Kernel integral operator, 010102 general mathematics, Covariance operator, 01 natural sciences, [STAT.ML] Statistics [stat]/Machine Learning [stat.ML], Kernel principal component analysis, Combinatorics, 010104 statistics & probability, Kernel method, [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], Fast convergence rates, Kernel spectrum estimation, Artificial Intelligence, Variable kernel density estimation, Kernel embedding of distributions, Polynomial kernel, String kernel, Kernel (statistics), Radial basis function kernel, Applied mathematics, Principal component regression, 0101 mathematics, Software, Mathematics
Abstract: The main goal of this paper is to prove inequalities on the reconstruction error for kernel principal component analysis. With respect to previous work on this topic, our contribution is twofold: (1) we give bounds that explicitly take into account the empirical centering step in this algorithm, and (2) we show that a "localized" approach allows to obtain more accurate bounds. In particular, we show faster rates of convergence towards the minimum reconstruction error; more precisely, we prove that the convergence rate can typically be faster than n ?1/2. We also obtain a new relative bound on the error. A secondary goal, for which we present similar contributions, is to obtain convergence bounds for the partial sums of the biggest or smallest eigenvalues of the kernel Gram matrix towards eigenvalues of the corresponding kernel operator. These quantities are naturally linked to the KPCA procedure; furthermore these results can have applications to the study of various other kernel algorithms. The results are presented in a functional analytic framework, which is suited to deal rigorously with reproducing kernel Hilbert spaces of infinite dimension.
Published: 2006
Full Text: View/download PDF

14. The 2011 Signal Separation Evaluation Campaign (SiSEC2011): - Audio source separation

Author: Francesco Nesta, Zbyněk Koldovský, Andreas Ziehe, Alexis Benichoux, Emmanuel Vincent, Shoko Araki, Guido Nolte, NTT Communication Science Laboratories, NTT Corporation, Fondazione Bruno Kessler [Trento, Italy] (FBK), Speech and sound data modeling and processing (METISS), Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria Rennes – Bretagne Atlantique, Institut National de Recherche en Informatique et en Automatique (Inria), Institute of Information Technology and Electronics (ITE), Technical University of Liberec, Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes 1 (UR1), and Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria Rennes – Bretagne Atlantique
Subjects: Computer science, Speech recognition, SIGNAL (programming language), Separation (aeronautics), 020206 networking & telecommunications, 02 engineering and technology, Blind signal separation, Non-negative matrix factorization, Task (project management), Speech enhancement, 030507 speech-language pathology & audiology, 03 medical and health sciences, [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, 0202 electrical engineering, electronic engineering, information engineering, Source separation, 0305 other medical science, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing
Abstract: International audience; This paper summarizes the audio part of the 2011 community-based Signal Separation Evaluation Campaign (SiSEC2011). Four speech and music datasets were contributed, including datasets recorded in noisy or dynamic environments and a subset of the SiSEC2010 datasets. The participants addressed one or more tasks out of four source separation tasks, and the results for each task were evaluated using different objective performance criteria. We provide an overview of the audio datasets, tasks and criteria.We also report the results achieved with the submitted systems, and discuss organization strategies for future campaigns.
Published: 2012

15. The Signal Separation Evaluation Campaign (2007-2010): Achievements and Remaining Challenges

Author: Alexey Ozerov, B. Vikrham Gowreesunker, Ngoc Q. K. Duong, Fabian J. Theis, Shoko Araki, Pau Bofill, Guido Nolte, Hiroshi Sawada, Dominik Lutter, Emmanuel Vincent, Speech and sound data modeling and processing (METISS), Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes 1 (UR1), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria Rennes – Bretagne Atlantique, Institut National de Recherche en Informatique et en Automatique (Inria), NTT Communication Science Laboratories, NTT Corporation, Institute of Bioinformatics and Systems Biology [München], Technische Universität Munchen - Université Technique de Munich [Munich, Allemagne] (TUM), Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Department of Computer Architecture [Barcelona] (DAC-UPC), Universitat Politècnica de Catalunya [Barcelona] (UPC), Texas Instruments - R&D center [USA], Texas Instruments, Institute of Biomathematics and Biometry - Helmholtz Zentrum München (IBB), Helmholtz-Zentrum München (HZM), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria Rennes – Bretagne Atlantique, and Helmholtz Zentrum München = German Research Center for Environmental Health
Subjects: Computer science, SIGNAL (programming language), 020206 networking & telecommunications, 02 engineering and technology, Latent variable, computer.software_genre, Data science, Field (computer science), [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, Control and Systems Engineering, Signal Processing, 0202 electrical engineering, electronic engineering, information engineering, Key (cryptography), Source separation, 020201 artificial intelligence & image processing, Computer Vision and Pattern Recognition, Data mining, Electrical and Electronic Engineering, computer, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Software
Abstract: International audience; We present the outcomes of three recent evaluation campaigns in the field of audio and biomedical source separation. These campaigns have witnessed a boom in the range of applications of source separation systems in the last few years, as shown by the increasing number of datasets from 1 to 9 and the increasing number of submissions from 15 to 34. We first discuss their impact on the definition of a reference evaluation methodology, together with shared datasets and software. We then present the key results obtained over almost all datasets. We conclude by proposing directions for future research and evaluation, based in particular on the ideas raised during the related panel discussion at the Ninth International Conference on Latent Variable Analysis and Signal Separation (LVA/ICA 2010).
Published: 2012
Full Text: View/download PDF

16. The 2011 Signal Separation Evaluation Campaign (SiSEC2011): - Biomedical Data Analysis

Author: Dominik Lutter, Alexis Benichoux, Zbyněk Koldovský, Guido Nolte, Andreas Ziehe, Emmanuel Vincent, Shoko Araki, Francesco Nesta, Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Institute of Biomathematics and Biometry - Helmholtz Zentrum München (IBB), Helmholtz-Zentrum München (HZM), Fondazione Bruno Kessler [Trento, Italy] (FBK), Speech and sound data modeling and processing (METISS), Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes 1 (UR1), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria Rennes – Bretagne Atlantique, Institut National de Recherche en Informatique et en Automatique (Inria), Institute of Information Technology and Electronics (ITE), Technical University of Liberec, NTT Communication Science Laboratories, NTT Corporation, Helmholtz Zentrum München = German Research Center for Environmental Health, Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), and Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria Rennes – Bretagne Atlantique
Subjects: 0303 health sciences, Computer science, Separation (statistics), SIGNAL (programming language), Bivariate analysis, computer.software_genre, Blind signal separation, Independent component analysis, Expression (mathematics), Task (project management), 03 medical and health sciences, 0302 clinical medicine, [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, Transfer entropy, Data mining, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, computer, 030217 neurology & neurosurgery, 030304 developmental biology
Abstract: International audience; This paper summarizes the bio part of the 2011 community based Signal Separation Evaluation Campaign (SiSEC2011). Two different data sets were given. In the first task, participants were asked to estimate the causal relations of underlying sources from simulated bivariate EEG data. In the second task, participants were asked to reconstruct signaling pathways or parts of it from the microarray expression profi les. The results for each task were evaluated using di fferent objective performance criteria. We provide an overview of the biomedical datasets, tasks and criteria, and we report on the achieved results.
Published: 2012
Full Text: View/download PDF

17. Ranking scientists and departments in a consistent manner

Author: Denis Bouyssou, Thierry Marchant, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Ghent University [Belgium] (UGENT), Universiteit Gent = Ghent University [Belgium] (UGENT), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
Subjects: Property (philosophy), Computer Networks and Communications, Computer science, IMPACT, Bibliometrics, 050905 science studies, ranking of scientists, SUCCESSIVE H-INDEXES, Consistency (negotiation), Artificial Intelligence, consistence, ranking of departments, ComputingMilieux_MISCELLANEOUS, Structure (mathematical logic), RESEARCH OUTPUT, ranking departments, 05 social sciences, BIBLIOMETRIC RANKINGS, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, Data science, Human-Computer Interaction, Mathematics and Statistics, Ranking, ranking scientists,ranking departments,consistence,Bibliometrics,ranking of scientists,ranking of departments, ranking scientists, 0509 other social sciences, 050904 information & library sciences, Citation, Software, Information Systems
Abstract: The standard data that we use when computing bibliometric rankings of scientists are their publication/ citation records, i.e., so many papers with 0 citation, so many with 1 citation, so many with 2 citations, etc. The standard data for bibliometric rankings of departments have the same structure. It is therefore tempting (and many authors gave in to temptation) to use the same method for computing rankings of scientists and rankings of departments. Depending on the method, this can yield quite surprising and unpleasant results. Indeed, with some methods, it may happen that the “best” department contains the “worst” scientists, and only them. This problem will not occur if the rankings satisfy a property called consistency, recently introduced in the literature. In this article, we explore the consequences of consistency and we characterize two families of consistent rankings. © 2011 Wiley Periodicals, Inc.
Published: 2011
Full Text: View/download PDF

18. Subjective expected utility without preferences

Author: Denis Bouyssou, Thierry Marchant, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
Subjects: Decision making under uncertainty, MODELS, Social Sciences, CATEGORIES, UNCERTAINTY, Ordered partitions, Von Neumann–Morgenstern utility theorem, Subjective Expected Utility,Conjoint Measurement, 050105 experimental psychology, AXIOMATIZATION, NONADDITIVE PROBABILITIES, PROSPECT-THEORY, Prospect theory, 0502 economics and business, Econometrics, 0501 psychology and cognitive sciences, Set (psychology), Representation (mathematics), ComputingMilieux_MISCELLANEOUS, General Psychology, Expected utility hypothesis, Axiom, 050205 econometrics, Mathematics, Conjoint measurement, RISK, Applied Mathematics, 05 social sciences, DECISION, Subjective expected utility, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, AXIOMS, Subjective Expected Utility, Probability distribution, Conjoint Measurement
Abstract: This paper proposes a theory of subjective expected utility based on primitives only involving the fact that an act can be judged either “attractive” or “unattractive”. We give conditions implying that there are a utility function on the set of consequences and a probability distribution on the set of states such that attractive acts have a subjective expected utility above some threshold. The numerical representation that is obtained has strong uniqueness properties.
Published: 2011
Full Text: View/download PDF

19. Multiattribute preference models with reference points

Author: Denis Bouyssou, Thierry Marchant, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
Subjects: Information Systems and Management, Theoretical computer science, General Computer Science, Relation (database), 0211 other engineering and technologies, Context (language use), 02 engineering and technology, Management Science and Operations Research, computer.software_genre, Industrial and Manufacturing Engineering, Set (abstract data type), Sugeno integral, 0202 electrical engineering, electronic engineering, information engineering, Representation (mathematics), Preference (economics), ComputingMilieux_MISCELLANEOUS, Mathematics, 021103 operations research, Axiomatic system, 16. Peace & justice, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, Modeling and Simulation, 020201 artificial intelligence & image processing, Data mining, Preference relation, computer
Abstract: In the context of multiple attribute decision making, preference models making use of reference points in an ordinal way have recently been introduced in the literature. This text proposes an axiomatic analysis of such models, with a particular emphasis on the case in which there is only one reference point. Our analysis uses a general conjoint measurement model resting on the study of traces induced on attributes by the preference relation and using conditions guaranteeing that these traces are complete. Models using reference points are shown to be a particular case of this general model. The number of reference points is linked to the number of equivalence classes distinguished by the traces. When there is only one reference point, the induced traces are quite rough, distinguishing at most two distinct equivalence classes. We study the relation between the model using a single reference point and other preference models proposed in the literature, most notably models based on concordance and models based on a discrete Sugeno integral.
Published: 2011
Full Text: View/download PDF

20. The Signal Separation Evaluation Campaign (2007-2010): Achievements and Remaining Challenges

Author: Vincent, Emmanuel, Araki, Shoko, Theis, Fabian J, Nolte, Guido, Bofill, Pau, Sawada, Hiroshi, Ozerov, Alexey, Gowreesunker, B. Vikrham, Lutter, Dominik, Duong, Ngoc, Speech and sound data modeling and processing (METISS), Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria Rennes – Bretagne Atlantique, Institut National de Recherche en Informatique et en Automatique (Inria), NTT Communication Science Laboratories, NTT Corporation, Institute of Biomathematics and Biometry - Helmholtz Zentrum München (IBB), Helmholtz Zentrum München = German Research Center for Environmental Health, Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Department of Computer Architecture [Barcelona] (DAC-UPC), Universitat Politècnica de Catalunya [Barcelona] (UPC), Texas Instruments - R&D center [USA], Texas Instruments, Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes 1 (UR1), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria Rennes – Bretagne Atlantique, and Helmholtz-Zentrum München (HZM)
Subjects: [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing
Abstract: La version 2 de ce rapport de recherche correspond à une rétractation, pour raisons scientifiques. Le fichier pdf reste disponible dans la version 1.; We present the outcomes of three recent evaluation campaigns in the field of audio and biomedical source separation. These campaigns have witnessed a boom in the range of applications of source separation systems in the last few years, as shown by the increasing number of datasets from 1 to 9 and the increasing number of submissions from 15 to 34. We first discuss their impact on the definition of a reference evaluation methodology, together with shared datasets and software. We then present the key results obtained over almost all datasets. We conclude by proposing directions for future research and evaluation, based in particular on the ideas raised during the related panel discussion at the Ninth International Conference on Latent Variable Analysis and Signal Separation (LVA/ICA 2010).
Published: 2011

21. Biorders with frontier

Author: Thierry Marchant, Denis Bouyssou, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
Subjects: Relation (database), MODELS, Semiorder, CATEGORIES, 02 engineering and technology, Interval (mathematics), PREFERENCES, SEMIORDERS, 01 natural sciences, Frontier, Combinatorics, 0202 electrical engineering, electronic engineering, information engineering, Interval order, 0101 mathematics, Representation (mathematics), INTRANSITIVE INDIFFERENCE, ComputingMilieux_MISCELLANEOUS, Mathematics, UTILITY, Algebra and Number Theory, INTERVAL ORDERS, 010102 general mathematics, Extension (predicate logic), [SHS.ECO]Humanities and Social Sciences/Economics and Finance, TIME, Mathematics and Statistics, Computational Theory and Mathematics, Product (mathematics), Biorder, 020201 artificial intelligence & image processing, Geometry and Topology, NUMERICAL REPRESENTATION
Abstract: This paper studies an extension of biorders that has a “frontier” between the relation and the absence of relation. This extension is motivated by a conjoint measurement problem consisting in the additive representation of ordered coverings defined on product sets of two components. We also investigate interval orders and semiorders with frontier.
Published: 2011
Full Text: View/download PDF

22. Bibliometric rankings of journals based on Impact Factors: An axiomatic approach

Author: Thierry Marchant, Denis Bouyssou, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
Subjects: Technology and Engineering, Journal rankings, Computer science, CITATION ANALYSIS, Decision theory, g-index, Library and Information Sciences, Bibliometrics, 050905 science studies, Computer Science::Digital Libraries, Citation analysis, HISTORY, Econometrics, TOOL, G-INDEX, GeneralLiterature_REFERENCE(e.g.,dictionaries,encyclopedias,glossaries), ComputingMilieux_MISCELLANEOUS, UTILITY, RISK, Impact Factor, CONSEQUENCES, Expected utility, Impact factor, 05 social sciences, Axiomatic system, SCIENCE, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, STANDS TODAY, Data science, Computer Science Applications, Ranking, Probability distribution, 0509 other social sciences, 050904 information & library sciences
Abstract: This paper proposes an axiomatic analysis of Impact Factors when used as tools for ranking journals. This analysis draws on the similarities between the problem of comparing distribution of citations among papers and that of comparing probability distributions on consequences as commonly done in decision theory. Our analysis singles out a number of characteristic properties of the ranking based on Impact Factors. We also suggest alternative ways of using distributions of citations to rank order journals.
Published: 2011
Full Text: View/download PDF

23. The 2010 Signal Separation Evaluation Campaign (SiSEC2010): - Audio source separation

Author: Dominik Lutter, B. Vikrham Gowreesunker, Fabian J. Theis, Alexey Ozerov, Ngoc Q. K. Duong, Hiroshi Sawada, Shoko Araki, Guido Nolte, NTT Communication Science Laboratories, NTT Corporation, Speech and sound data modeling and processing (METISS), Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria Rennes – Bretagne Atlantique, Institut National de Recherche en Informatique et en Automatique (Inria), Texas Instruments - R&D center [USA], Texas Instruments, Institute of Biomathematics and Biometry - Helmholtz Zentrum München (IBB), Helmholtz Zentrum München = German Research Center for Environmental Health, Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes 1 (UR1), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria Rennes – Bretagne Atlantique, and Helmholtz-Zentrum München (HZM)
Subjects: 0303 health sciences, Computer science, Speech recognition, Separation (aeronautics), Blind signal separation, Signal, Task (project management), Non-negative matrix factorization, Speech enhancement, 030507 speech-language pathology & audiology, 03 medical and health sciences, [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, Source separation, 0305 other medical science, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, 030304 developmental biology
Abstract: International audience; This paper introduces the audio part of the 2010 community-based Signal Separation Evaluation Campaign (SiSEC2010). Seven speech and music datasets were contributed, which include datasets recorded in noisy or dynamic environments, in addition to the SiSEC2008 datasets. The source separation problems were split into five tasks, and the results for each task were evaluated using different objective performance criteria. We provide an overview of the audio datasets, tasks and criteria. We also report the results achieved with the submitted systems, and discuss organization strategies for future campaigns.
Published: 2010

24. The 2010 Signal Separation Evaluation Campaign (SiSEC2010): - Biomedical source separation

Author: Araki, Shoko, Theis, Fabian J, Nolte, Guide, Lutter, Dominik, Ozerov, Alexey, Gowreesunker, B. Vikrham, Sawada, Hiroshi, Duong, Ngoc, NTT Communication Science Laboratories, NTT Corporation, Institute of Biomathematics and Biometry - Helmholtz Zentrum München (IBB), Helmholtz Zentrum München = German Research Center for Environmental Health, Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Speech and sound data modeling and processing (METISS), Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria Rennes – Bretagne Atlantique, Institut National de Recherche en Informatique et en Automatique (Inria), Texas Instruments - R&D center [USA], Texas Instruments, Helmholtz-Zentrum München (HZM), Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes 1 (UR1), and Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria Rennes – Bretagne Atlantique
Subjects: [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing
Abstract: International audience; We present an overview of the biomedical part of the 2010 community-based Signal Separation Evaluation Campaign (SiSEC2010), coordinated by the authors. In addition to the audio tasks which have been evaluated in the previous SiSEC, SiSEC2010 considered several biomedical tasks. Here, three biomedical datasets from molecular biology (gene expression profiles) and neuroscience (EEG) were contributed. This paper describes the biomedical datasets, tasks and evaluation criteria. This paper also reports the results of the biomedical part of SiSEC2010 achieved by participants.
Published: 2010

25. Additive conjoint measurement with ordered categories

Author: Denis Bouyssou, Thierry Marchant, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), and Ghent University [Belgium] (UGENT)
Subjects: Information Systems and Management, Theoretical computer science, General Computer Science, 0211 other engineering and technologies, 02 engineering and technology, Management Science and Operations Research, 050105 experimental psychology, Industrial and Manufacturing Engineering, Additive function, 0501 psychology and cognitive sciences, Uniqueness, Representation (mathematics), Additive model, Preference (economics), ComputingMilieux_MISCELLANEOUS, Mathematics, 021103 operations research, Binary relation, business.industry, 05 social sciences, Sorting, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, Modeling and Simulation, Artificial intelligence, Preference relation, business
Abstract: Conjoint measurement studies binary relations defined on product sets and investigates the existence and uniqueness of, usually additive, numerical representations of such relations. It has proved to be quite a powerful tool to analyze and compare MCDM techniques designed to build a preference relation between multiattributed alternatives and has been an inspiring guide to many assessment protocols. The aim of this paper is to show that additive representations can be obtained on the basis of much poorer information than a preference relation. We will suppose here that the decision maker only specifies for each object if he/she finds it “attractive” (better than the status quo), “unattractive” (worse than the status quo) or “neutral” (equivalent to the status quo). We show how to build an additive representation, with tight uniqueness properties, using such an ordered partition of the set of objects. On a theoretical level, this paper shows that classical results of additive conjoint measurement can be extended to cover the case of ordered partitions and wishes to be a contribution to the growing literature on the foundations of sorting techniques in MCDM. On a more practical level, our results suggest an assessment strategy of an additive model on the basis of an ordered partition.
Published: 2010
Full Text: View/download PDF

26. Some nonasymptotic results on resampling in high dimension, II: Multiple tests

Author: Gilles Blanchard, Etienne Roquain, Sylvain Arlot, Laboratoire d'informatique de l'école normale supérieure (LIENS), École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Institut für Mathematik [Potsdam], Universität Potsdam, Weierstraß-Institut für Angewandte Analysis und Stochastik = Weierstrass Institute for Applied Analysis and Stochastics [Berlin] (WIAS), Forschungsverbund Berlin e.V. (FVB) (FVB), Laboratoire de Probabilités et Modèles Aléatoires (LPMA), Centre National de la Recherche Scientifique (CNRS)-Université Paris Diderot - Paris 7 (UPD7)-Université Pierre et Marie Curie - Paris 6 (UPMC), Département d'informatique - ENS Paris (DI-ENS), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS), École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Paris (ENS-PSL), University of Potsdam = Universität Potsdam, and Université Pierre et Marie Curie - Paris 6 (UPMC)-Université Paris Diderot - Paris 7 (UPD7)-Centre National de la Recherche Scientifique (CNRS)
Subjects: Statistics and Probability, Clustering high-dimensional data, Family-wise error, Multivariate random variable, Context (language use), 01 natural sciences, nonasymptotic error control, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], 010104 statistics & probability, Dimension (vector space), [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], resampling, [MATH.MATH-ST]Mathematics [math]/Statistics [math.ST], Resampling, resampled quantile, 62G09, 0502 economics and business, Statistics, Applied mathematics, 0101 mathematics, 050205 econometrics, Mathematics, Confidence region, Covariance matrix, multiple testing, 05 social sciences, high-dimensional data, Multiple comparisons problem, Statistics, Probability and Uncertainty, 62G10
Abstract: International audience; In the context of correlated multiple tests, we aim to nonasymptotically control the family-wise error rate (FWER) using resampling-type procedures. We observe repeated realizations of a Gaussian random vector in possibly high dimension and with an unknown covariance matrix, and consider the one- and two-sided multiple testing problem for the mean values of its coordinates. We address this problem by using the confidence regions developed in the companion paper [Ann. Statist. 38 (2010) 51-82], which lead directly to single-step procedures; these can then be improved using step-down algorithms, following an established general methodology laid down by Romano and Wolf [J. Amer. Statist. Assoc. 100 (2005) 94–108]. This gives rise to several different procedures, whose performances are compared using simulated data.
Published: 2010
Full Text: View/download PDF

27. Consistent bibliometric rankings of authors and of journals

Author: Denis Bouyssou, Thierry Marchant, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Ghent University [Belgium] (UGENT), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
Subjects: Ranking of journals, Computer science, IMPACT, media_common.quotation_subject, InformationSystems_INFORMATIONSTORAGEANDRETRIEVAL, Citations, Library and Information Sciences, Bibliometrics, 050905 science studies, Computer Science::Digital Libraries, HIRSCH-INDEX, Ranking of authors, Quality (business), Impact factor, GeneralLiterature_REFERENCE(e.g.,dictionaries,encyclopedias,glossaries), Axiom, ComputingMilieux_MISCELLANEOUS, media_common, 05 social sciences, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, Data science, Computer Science Applications, INDIVIDUALS, Mathematics and Statistics, Consistent rankings, 0509 other social sciences, 050904 information & library sciences
Abstract: Rankings of journals and rankings of scientists are usually discussed separately. We argue that a consistent approach to both rankings is desirable because both the quality of a journal and the quality of a scientist depend on the papers it/he publishes. We present a pair of consistent rankings (impact factor for the journals and total number of citations for the authors) and we provide an axiomatic characterization thereof.
Published: 2010
Full Text: View/download PDF

28. Some non-asymptotic results on resampling in high dimension, I: Confidence regions, II: Multiple tests

Author: Arlot, Sylvain, Blanchard, Gilles, Roquain, Etienne, Laboratoire d'informatique de l'école normale supérieure (LIENS), École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Models of visual object recognition and scene understanding (WILLOW), Département d'informatique de l'École normale supérieure (DI-ENS), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria Paris-Rocquencourt, Institut National de Recherche en Informatique et en Automatique (Inria), Weierstraß-Institut für Angewandte Analysis und Stochastik = Weierstrass Institute for Applied Analysis and Stochastics [Berlin] (WIAS), Forschungsverbund Berlin e.V. (FVB) (FVB), Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Laboratoire de Probabilités et Modèles Aléatoires (LPMA), Centre National de la Recherche Scientifique (CNRS)-Université Paris Diderot - Paris 7 (UPD7)-Université Pierre et Marie Curie - Paris 6 (UPMC), This work was supported in part by the IST and ICT Programmes of the European Community, successively under the PASCAL (IST-2002-506778) and PASCAL2 (ICT-216886) networks of excellence., Département d'informatique - ENS Paris (DI-ENS), École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS), Université Pierre et Marie Curie - Paris 6 (UPMC)-Université Paris Diderot - Paris 7 (UPD7)-Centre National de la Recherche Scientifique (CNRS), Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL), and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Inria Paris-Rocquencourt
Subjects: Statistics::Theory, confidence regions, family-wise error, multiple testing, high dimensional data, non-asymptotic error control, Mathematics - Statistics Theory, Statistics Theory (math.ST), [STAT.TH]Statistics [stat]/Statistics Theory [stat.TH], cross-validation, AMS classification: Primary 62G15, secondary 62G09, 62G10, 62G15 (Primary) 62G09 (Secondary), 62G10 (Primary) 62G09 (Secondary), resampling, concentration inequalities, [MATH.MATH-ST]Mathematics [math]/Statistics [math.ST], resampled quantile, FOS: Mathematics
Abstract: We study generalized bootstrap confidence regions for the mean of a random vector whose coordinates have an unknown dependency structure. The random vector is supposed to be either Gaussian or to have a symmetric and bounded distribution. The dimensionality of the vector can possibly be much larger than the number of observations and we focus on a nonasymptotic control of the confidence level, following ideas inspired by recent results in learning theory. We consider two approaches, the first based on a concentration principle (valid for a large class of resampling weights) and the second on a resampled quantile, specifically using Rademacher weights. Several intermediate results established in the approach based on concentration principles are of interest in their own right. We also discuss the question of accuracy when using Monte Carlo approximations of the resampled quantities., Published in at http://dx.doi.org/10.1214/08-AOS667; http://dx.doi.org/10.1214/08-AOS668 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)
Published: 2010
Full Text: View/download PDF

29. Ordered categories and additive conjoint measurement on connected sets

Author: Thierry Marchant, Denis Bouyssou, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
Subjects: Discrete mathematics, 021103 operations research, Binary relation, Applied Mathematics, 05 social sciences, 0211 other engineering and technologies, 02 engineering and technology, Cartesian product, 16. Peace & justice, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, 050105 experimental psychology, Combinatorics, symbols.namesake, symbols, Partition (number theory), 0501 psychology and cognitive sciences, Additive model, General Psychology, ComputingMilieux_MISCELLANEOUS, Mathematics
Abstract: Suppose that a binary relation is given on a n-fold Cartesian product. The study of the conditions guaranteeing the existence of n value functions such that the binary relation can be additively represented is known as additive conjoint measurement. In this paper we analyze a related problem: given a partition of a Cartesian product into r ordered categories, what conditions do ensure the representability of the partition in an additive model?
Published: 2009
Full Text: View/download PDF

30. Additive and decomposable conjoint measurement with ordered categories

Author: Bouyssou, Denis, Marchant, Thierry, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL), Research Group on Combinatorial Algorithms and Algorithmic Graph Theory (Ghent University), Universiteit Gent = Ghent University [Belgium] (UGENT), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), and Department of Data Analysis
Subjects: Conjoint measurement, UTADIS, Sorting, [INFO.INFO-RO]Computer Science [cs]/Operations Research [cs.RO], [INFO.INFO-DM]Computer Science [cs]/Discrete Mathematics [cs.DM], [SHS.ECO]Humanities and Social Sciences/Economics and Finance, Decision with multiple attributes
Abstract: Conjoint measurement studies binary relations defined on product sets and investigates the existence and uniqueness of numerical representations of such relations. It has proved to be quite a powerful tool to analyze and compare MCDM techniques designed to build a preference relation between multiattributed alternatives and has been an inspiring guide to many assessment protocols. These MCDM techniques lead to a relative evaluation model of the alternatives through a preference relation. Such models are not always appropriate to build meaningful recommendations. This has recently lead to the development of MCDM techniques aiming at building evaluation models having a more absolute character. In such techniques, the output of the analysis is, most often, a partition of the set of alternatives into several ordered categories defined with respect to outside norms, e.g., separating “Attractive” and “Unattractive” alternatives. In spite of their interest, the theoretical foundations of such MCDM techniques have not been much investigated. The purpose of this paper is to contribute to this analysis. More precisely, we show how to adapt classic conjoint measurement results to make them applicable for the study of such MCDM techniques. We concentrate on additive models. Our results may be seen as an attempt to provide an axiomatic basis to the well-known UTADIS technique that sorts alternatives using an additive value function model. Keywords: Decision with multiple attributes, Sorting, Conjoint measurement, UTADIS. La théorie du mesurage conjoint étudie la question de la représentation numérique d'une relation binaire définie sur un produit cartésien. Cette théorie s'est révélée très utile pour comparer et analyser diverses techniques d'aide multicritère à la décision. Elle a également été la source de nombreux protocoles d'élicitation. Les techniques d'aide à la décision utilisant une relation binaire comparant des actions évaluées sur plusieurs attributs conduisent, en général, à des modèles d'évaluation ayant un caractère relatif. Or, de tels modèles ne sont pas toujours adaptés pour bâtir une prescription pertinente. Ceci a conduit au développement de techniques multicritères conduisant à des modèles d'évaluation ayant un caractère plus absolu. Dans ces techniques, le résultat se présente généralement sous la forme d'une aectation des actions à diverses catégories ordonnées, ces catégories étant définies par rapport à des normes indépendantes des actions à évaluer. On pourra, par exemple, séparer les actions satisfaisantes de celles étant insatisfaisantes. En dépit de leur intérêt, les fondements théoriques de telles méthodes ont été peu étudiés. L'objectif de cet article est de contribuer à cette étude. Plus précisément, on montre comment adapter les résultats classiques du mesurage conjoint pour couvrir le cas de catégories ordonnées. On étudie plus spécifiquement le cas de représentations additives. Ce travail peut alors être vu comme une tentative de donner à la méthode UTADIS une base théorique solide. Mots-clés: Analyse multicritère, Tri, Mesurage conjoint, UTADIS.
Published: 2008

31. Two simple sufficient conditions for FDR control

Author: Gilles Blanchard, Etienne Roquain, Laboratoire de Probabilités et Modèles Aléatoires (LPMA), Université Pierre et Marie Curie - Paris 6 (UPMC)-Université Paris Diderot - Paris 7 (UPD7)-Centre National de la Recherche Scientifique (CNRS), Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Centre National de la Recherche Scientifique (CNRS)-Université Paris Diderot - Paris 7 (UPD7)-Université Pierre et Marie Curie - Paris 6 (UPMC), This work was supported in part by the IST Programme of the European Community, under the PASCAL Network of Excellence, IST-2002-506778., and Publica
Subjects: Statistics and Probability, False discovery rate, Theoretical computer science, Dependency (UML), Mathematics - Statistics Theory, Statistics Theory (math.ST), Mathematical proof, 01 natural sciences, step-down, 010104 statistics & probability, PRDS condition, [MATH.MATH-ST]Mathematics [math]/Statistics [math.ST], 0502 economics and business, FOS: Mathematics, 0101 mathematics, Control (linguistics), step-up-down, GeneralLiterature_REFERENCE(e.g.,dictionaries,encyclopedias,glossaries), Independence (probability theory), ComputingMilieux_MISCELLANEOUS, 050205 econometrics, Mathematics, Simple (philosophy), weighted p-values, False Discovery Rate, step-up, multiple testing, 05 social sciences, [STAT.TH]Statistics [stat]/Statistics Theory [stat.TH], 62J15, 62G10, [MATH.MATH-PR]Mathematics [math]/Probability [math.PR], Multiple comparisons problem, 62J15, Uncountable set, Statistics, Probability and Uncertainty, 62G10
Abstract: We show that the control of the false discovery rate (FDR) for a multiple testing procedure is implied by two coupled simple sufficient conditions. The first one, which we call ``self-consistency condition'', concerns the algorithm itself, and the second, called ``dependency control condition'' is related to the dependency assumptions on the $p$-value family. Many standard multiple testing procedures are self-consistent (e.g. step-up, step-down or step-up-down procedures), and we prove that the dependency control condition can be fulfilled when choosing correspondingly appropriate rejection functions, in three classical types of dependency: independence, positive dependency (PRDS) and unspecified dependency. As a consequence, we recover earlier results through simple and unifying proofs while extending their scope to several regards: weighted FDR, $p$-value reweighting, new family of step-up procedures under unspecified $p$-value dependency and adaptive step-up procedures. We give additional examples of other possible applications. This framework also allows for defining and studying FDR control for multiple testing procedures over a continuous, uncountable space of hypotheses., Comment: Published in at http://dx.doi.org/10.1214/08-EJS180 the Electronic Journal of Statistics (http://www.i-journals.org/ejs/) by the Institute of Mathematical Statistics (http://www.imstat.org)
Published: 2008
Full Text: View/download PDF

32. Resampling-based confidence regions and multiple tests for a correlated random vector

Author: Arlot, Sylvain, Blanchard, Gilles, Roquain, Etienne, Laboratoire de Mathématiques d'Orsay (LM-Orsay), Université Paris-Sud - Paris 11 (UP11)-Centre National de la Recherche Scientifique (CNRS), Fraunhofer FIRST, IDA group (FHG FIRST.IDA), Fraunhofer FIRST-Fraunhofer-Gesellschaft, Unité Mathématique Informatique et Génome (MIG), Institut National de la Recherche Agronomique (INRA), Nader H. Bshouty and Claudio Gentile, Model selection in statistical learning (SELECT), Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de Mathématiques d'Orsay (LMO), Université Paris-Sud - Paris 11 (UP11)-Centre National de la Recherche Scientifique (CNRS)-Université Paris-Sud - Paris 11 (UP11)-Centre National de la Recherche Scientifique (CNRS), Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Laboratoire de Mathématiques d'Orsay (LMO), Université Paris-Sud - Paris 11 (UP11)-Centre National de la Recherche Scientifique (CNRS)-Université Paris-Sud - Paris 11 (UP11)-Centre National de la Recherche Scientifique (CNRS)-Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), and Centre National de la Recherche Scientifique (CNRS)-Université Paris-Sud - Paris 11 (UP11)
Subjects: resampling, gaussian vector, concentration inequalities, [MATH.MATH-ST]Mathematics [math]/Statistics [math.ST], multiple testing, unknown correlations, resampled quantiles, confidence region, MSC : 62G09, 62H15, cross-validation
Abstract: submitted to COLT; We derive non-asymptotic confidence regions for the mean of a random vector whose coordinates have an unknown dependence structure. The random vector is supposed to be either Gaussian or to have a symmetric bounded distribution, and we observe $n$ i.i.d copies of it. The confidence regions are built using a data-dependent threshold based on a weighted bootstrap procedure. We consider two approaches, the first based on a concentration approach and the second on a direct boostrapped quantile approach. The first one allows to deal with a very large class of resampling weights while our results for the second are restricted to Rademacher weights. However, the second method seems more accurate in practice. Our results are motivated by multiple testing problems, and we show on simulations that our procedures are better than the Bonferroni procedure (union bound) as soon as the observed vector has sufficiently correlated coordinates.
Published: 2007
Full Text: View/download PDF

33. An axiomatic approach to noncompensatory sorting methods in MCDM, II: More than two categories

Author: Thierry Marchant, Denis Bouyssou, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
Subjects: 021103 operations research, Information Systems and Management, Sorting algorithm, General Computer Science, Operations research, 0211 other engineering and technologies, Sorting, Axiomatic system, 02 engineering and technology, Management Science and Operations Research, Multiple-criteria decision analysis, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, Industrial and Manufacturing Engineering, Modeling and Simulation, 0202 electrical engineering, electronic engineering, information engineering, Feature (machine learning), sort, 020201 artificial intelligence & image processing, ELECTRE, Mathematical economics, Axiom, ComputingMilieux_MISCELLANEOUS, Mathematics
Abstract: A number of recent papers have investigated the foundations of methods allowing to sort multi-attributed alternatives between several ordered categories. This paper has a similar objective. Our analysis uses a general conjoint measurement framework, encompassing most sorting models used in MDCM, that was proposed in the literature. Within this framework, we provide an axiomatic analysis of what we call noncompensatory sorting models, with or without veto effects. These noncompensatory sorting models contain the pessimistic version of ELECTRE TRI as a particular case. Our analysis can be seen as an attempt to give a firm axiomatic basis to ELECTRE TRI, while emphasizing its specific feature, i.e., the rather poor information that this model uses on each attribute.
Published: 2007
Full Text: View/download PDF

34. A conjoint measurement approach to the discrete Sugeno integral

Author: Bouyssou, Denis, Marchant, Thierry, Pirlot, Marc, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Faculté polytechnique de Mons, and Université de Mons (UMons)
Subjects: [INFO.INFO-CC]Computer Science [cs]/Computational Complexity [cs.CC], conjoint measurement, Sugeno integral, [INFO.INFO-RO]Computer Science [cs]/Operations Research [cs.RO], MCDM
Abstract: A note on a result of Greco, Matarazzo and Slowinski; In a recent paper (European Journal of Operational Research, 158, 271–292, 2004), S. Greco, B. Matarazzo and R. Słowinski have stated without proof a result characterizing binary relations on product sets that can be represented using a discrete Sugeno integral. To our knowledge, this is the first result about a fuzzy integral that applies to non-necessarily homogeneous product sets and only uses a binary relation on this set as a primitive. This is of direct interest to MCDM. The main purpose of this note is to propose a proof of this important result. Thereby, we study the connections between the discrete Sugeno integral and a non-numerical model called the noncompensatory model. We also show that the main condition used in the result of S. Greco, B. Matarazzo and R. Słowi´nski can be factorized in such a way that the discrete Sugeno integral model can be viewed as a particular case of a general decomposable representation.
Published: 2006

35. Discussion: Local Rademacher complexities and oracle inequalities in risk minimization

Author: Gilles Blanchard, Pascal Massart, Publica, Laboratoire de Mathématiques d'Orsay (LMO), Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), and Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft)
Subjects: Statistics and Probability, 02 engineering and technology, 01 natural sciences, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], 010104 statistics & probability, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], [MATH.MATH-ST]Mathematics [math]/Statistics [math.ST], 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Minification, 0101 mathematics, Statistics, Probability and Uncertainty, Mathematical economics, ComputingMilieux_MISCELLANEOUS, Oracle inequality, Mathematics
Abstract: Discussion of ``2004 IMS Medallion Lecture: Local Rademacher complexities and oracle inequalities in risk minimization'' by V. Koltchinskii [arXiv:0708.0083]
Published: 2006
Full Text: View/download PDF

36. Occam's hammer: a link between randomized learning and multiple testing FDR control

Author: Blanchard, Gilles, Fleuret, François, Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Laboratoire de Mathématiques d'Orsay (LMO), Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), Centre National de la Recherche Scientifique (CNRS), Computer Vision Laboratory [EPFL] (CVLAB), Ecole Polytechnique Fédérale de Lausanne (EPFL), IDIAP Research Institute, and Nader Bshouty and Claudio Gentile
Subjects: FOS: Computer and information sciences, Computer Science - Learning, False Discovery Rate (FDR), [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], learning theory, [MATH.MATH-ST]Mathematics [math]/Statistics [math.ST], multiple testing, FOS: Mathematics, Mathematics - Statistics Theory, PAC-Bayes, Statistics Theory (math.ST), randomized estimation, Machine Learning (cs.LG)
Abstract: We establish a generic theoretical tool to construct probabilistic bounds for algorithms where the output is a subset of objects from an initial pool of candidates (or more generally, a probability distribution on said pool). This general device, dubbed "Occam's hammer'', acts as a meta layer when a probabilistic bound is already known on the objects of the pool taken individually, and aims at controlling the proportion of the objects in the set output not satisfying their individual bound. In this regard, it can be seen as a non-trivial generalization of the "union bound with a prior'' ("Occam's razor''), a familiar tool in learning theory. We give applications of this principle to randomized classifiers (providing an interesting alternative approach to PAC-Bayes bounds) and multiple testing (where it allows to retrieve exactly and extend the so-called Benjamini-Yekutieli testing procedure)., Comment: 13 pages -- conference communication type format
Published: 2006
Full Text: View/download PDF

37. A simple Bayes factor for testing measurement-theoretic axioms

Author: Marchant, Thierry, Hugonnard-Roche, Marie-Hélène, Department of Data Analysis, and Universiteit Gent = Ghent University [Belgium] (UGENT)
Subjects: [INFO.INFO-CC]Computer Science [cs]/Computational Complexity [cs.CC], [INFO.INFO-RO] Computer Science [cs]/Operations Research [cs.RO], empirical test, [INFO.INFO-CC] Computer Science [cs]/Computational Complexity [cs.CC], [INFO.INFO-RO]Computer Science [cs]/Operations Research [cs.RO], Measurement theory
Abstract: A lot of models used to represent preferences or other relations have been characterized in the framework of measurement theory. To assess the empirical validity of one of these models, we can either test the model itself or test the axioms characterizing it. When a model is rejected, the latter approach has the advantage of indicating why it is rejected. The available statistical techniques for testing measurementtheoretic axioms are complex and not always satisficing. In this paper, we present a new statistical technique based on a Bayes factor, very simple and flexible.
Published: 2006

38. Evaluation and Decision Models with Multiple Criteria

Author: Bouyssou, Denis, Marchant, Thierry, Pirlot, Marc, Tsoukiàs, Alexis, Vincke, Philippe, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, Universiteit Gent = Ghent University [Belgium] (UGENT), Université de Mons (UMons), CODE (CODE), Université libre de Bruxelles (ULB), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)
Subjects: 021105 building & construction, 0211 other engineering and technologies, 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, 02 engineering and technology, [SHS.ECO]Humanities and Social Sciences/Economics and Finance, ComputingMilieux_MISCELLANEOUS
Abstract: International audience
Published: 2006
Full Text: View/download PDF

39. An axiomatic approach to noncompensatory sorting methods in MCDM I: The case of two categories (juin 2005)

Author: Bouyssou, Denis, Marchant, Thierry, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, and Universiteit Gent = Ghent University [Belgium] (UGENT)
Subjects: Conjoint measurement, ELECTRE TRI, [STIC]domain_stic, Sorting, Noncompensatory sorting methods, Decision with multiple attributes
Abstract: In the literature on MCDM, many methods have been proposed in order to sort alternatives evaluated on several attributes into ordered categories. Most of them were proposed on an ad hoc basis. The purpose of this paper is to contribute to a recent trend of research aiming at giving these methods sound theoretical foundations. Using tools from conjoint measurement, we provide an axiomatic analysis of the partitions of alternatives into two categories that can be obtained using what we call “noncompensatory sorting models”. These models have strong links with the pessimistic version of ELECTRE TRI. Our analysis allows to pinpoint what appears to be the main distinctive features of ELECTRE TRI when compared to other sorting methods. It also gives hints on the various methods that have been proposed to assess the parameters of ELECTRE TRI on the basis of assignment examples.
Published: 2005

40. An axiomatic approach to noncompensatory sorting methods in MCDM II: More than two categories (janvier 2005, révisé octobre 2005)

Author: Bouyssou, Denis, Marchant, Thierry, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Department of Data Analysis, and Universiteit Gent = Ghent University [Belgium] (UGENT)
Subjects: Conjoint measurement, ELECTRE TRI, [STIC]domain_stic, Sorting models, Noncompensation, Decision with multiple attributes
Abstract: A number of recent papers have investigated the foundations of methods allowing to sort multi-attributed alternatives between several ordered categories. This paper has a similar objective. Our analysis uses a general conjoint measurement framework, encompassing most sorting models used in MDCM, that was proposed in the literature. Within this framework, we provide an axiomatic analysis of what we call noncompensatory sorting models, with or without veto eﬀects. These noncompensatory sorting models contain the pessimistic version of ELECTRE TRI as a particular case. Our analysis can be seen as an attempt to give a ﬁrm axiomatic basis to ELECTRE TRI, while emphasizing its speciﬁc feature, i.e., the rather poor information that this model uses on each attribute.
Published: 2005

41. Kernel Projection Machine: selection de modeles pour ce nouvel algorithme de classification

Author: Zwald, Laurent, Blanchard, Gilles, Zwald, Laurent, Laboratoire de Mathématiques d'Orsay (LM-Orsay), Centre National de la Recherche Scientifique (CNRS)-Université Paris-Sud - Paris 11 (UP11), Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), and Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft)
Subjects: ComputingMethodologies_PATTERNRECOGNITION, [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], [STAT.ML] Statistics [stat]/Machine Learning [stat.ML]
Abstract: International audience; This work presents a theoretical study of the Kernel Projection Machine, a classification algorithm which is an alternate to the Support Vector Machine (SVM). We propose to replace the Tikhonov regularization in the SVM by a penalized finite-dimensional projection. We show that a penalty term proportional to the dimension is appropriate under a ``gap'' hypothesis on the ditribution of class given observation. We also show how to apply this algorithm in practice and present examples of numerical results.
Published: 2005

42. On the Convergence of Eigenspaces in Kernel Principal Component Analysis

Author: Zwald, Laurent, Blanchard, Gilles, Département de Mathématiques [ORSAY], Université Paris-Sud - Paris 11 (UP11), Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), and Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft)
Subjects: [STAT.ML]Statistics [stat]/Machine Learning [stat.ML]
Abstract: International audience; This paper presents a non-asymptotic statistical analysis of Kernel-PCA with a focus different from the one proposed in previous work on this topic (\cite{ShaWilCriKan02CDL}, chapter \ref{KPCA1chap}). Here instead of considering the reconstruction error of KPCA we are interested in approximation error bounds for the eigenspaces themselves. We prove an upper bound depending on the spacing between eigenvalues but not on the dimensionality of the eigenspace. As a consequence this allows to infer stability results for these estimated spaces.
Published: 2005

43. Kernel Projection Machine: a New Tool for Pattern Recognition

Author: Blanchard, Gilles, Massart, Pascal, Vert, Régis, Zwald, Laurent, Department Intelligent Data Analysis [Berlin] (Fraunhofer FIRST.IDA), Fraunhofer Institute for Manufacturing Engineering and Automation (Fraunhofer IPA), Fraunhofer (Fraunhofer-Gesellschaft)-Fraunhofer (Fraunhofer-Gesellschaft), Laboratoire de Mathématiques d'Orsay (LM-Orsay), Centre National de la Recherche Scientifique (CNRS)-Université Paris-Sud - Paris 11 (UP11), Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Zwald, Laurent
Subjects: [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], [STAT.ML] Statistics [stat]/Machine Learning [stat.ML]
Abstract: International audience; This paper investigates the effect of Kernel Principal Component Analysis (KPCA) within the classification framework, essentially the regularization properties of this dimensionality reduction method. KPCA has been previously used as a pre-processing step before applying an SVM but we point out that this method is somewhat redundant from a regularization point of view and we propose a new algorithm called Kernel Projection Machine to avoid this redundancy, based on an analogy with the statistical framework of regression for a Gaussian white noise model. Preliminary experimental results show that this algorithm reaches the same performances as an SVM.
Published: 2004

44. Independent Genetic Mapping Experiments Identify Diverse Molecular Determinants of Host Adaptation in a Generalist Herbivore.

Author: Villacis-Perez E, De Graeve F, De Beer B, Ali Alshami S, De Jong R, De Meyer T, and Van Leeuwen T
Abstract: Interactions between plants and herbivores promote evolutionary change. Studying the evolution of herbivore mechanisms aimed to cope with different host plant species is a critical intersection between evolutionary biology and sustainable pest management. Generalist herbivores are of particular interest, as hybridization between genetically distinct populations can increase the standing genetic variation and therefore the adaptive potential of the species. Tetranychus urticae is a generalist arthropod known for its adaptive potential, evidenced in its immense host range and ability to develop metabolic resistance to xenobiotics. However, the molecular underpinnings associated with the potential of host adaptation and the consequences of host adaptation in this, and many other pests remain elusive. Here, we use two independent, empirical approaches to identify and map the genetic basis of host plant performance and adaptation in genetically distinct populations of T. urticae. In the first approach, we subject a genetically diverse mite population to tomato selection and map genomic regions linked to the phenotypic evolution of increased reproductive performance. In the second approach, we map genomic regions responsible for performance on tomato by comparing the genomes of pooled individuals from an F2 backcross between populations with high and low reproductive performance. Both approaches revealed specific and shared genomic regions associated with host plant performance and adaptation and key candidate genes were identified. Our findings highlight the power of spider mite genetic approaches to identify the complex genetic basis of host adaptation in generalist herbivores., (© 2024 John Wiley & Sons Ltd.)
Published: 2024
Full Text: View/download PDF

45. Low thrombin inactivation capacity is associated with an increased risk of recurrent ischemic events after ischemic stroke at a young age.

Author: Spiegelenberg JP, De Laat-Kremers R, Roest M, de Laat B, van Gelder MMHJ, Tuladhar AM, Middeldorp S, de Leeuw FE, and Leentjens J
Abstract: Background: Patients with ischemic stroke at a young age (18-50 years) have an increased long-term risk of recurrent ischemic events. Hypercoagulability may contribute to this high risk., Objectives: To investigate associations between in and ex vivo hemostatic parameters and recurrent ischemic events after an ischemic stroke or transient ischemic attack at a young age., Methods: We included patients with ischemic stroke or TIA between 1980 and 2010 from the prospective FUTURE cohort. Blood samples were collected in 2010 and patients were followed for recurrent ischemic events from 2010 to 2023. Pro- and anticoagulant markers and thrombin generation assay were measured. Thrombin dynamic analysis was used to study underlying pro- and anticoagulant processes. Hazard ratios (HR) per standard deviation increase were assessed with cause-specific hazard models., Results: Of the initial cohort of 581 patients, 332 were eligible. The median time between the index event and 2010 was 7.6 years. During a mean follow-up of 6.5 years, 70 of 332 (21.1%) patients experienced a recurrent ischemic event. Lower antithrombin levels (adjusted HR=0.77; 95% CI 0.60-0.98) and higher fibrinogen levels (HR=1.35; CI 1.04-1.73) were associated with higher risk of recurrent ischemic events. Plasma thrombin generation was not associated with recurrence. However, the thrombin decay constant (HR=0.67; CI 0.51-0.87) was associated with a lower risk of recurrent ischemic events., Conclusion: After an ischemic stroke or TIA at a young age, thrombin decay constant, reflecting reduced protection against thrombin (low antithrombin) and decreased potential to inhibit thrombin (high fibrinogen), is associated with recurrent ischemic events., (Copyright © 2024. Published by Elsevier Inc.)
Published: 2024
Full Text: View/download PDF

46. Modelling the effect of base component properties and processing conditions on mixture products using probabilistic, knowledge-guided neural networks.

Author: Borja M, Dhondt J, Bertels J, Van Hauwermeiren D, and Verwaeren J
Abstract: Development of materials by mixing different base components is a widespread methodology to create materials with improved properties compared to those of its base components. However, efficient determination of the properties of mixture-based materials during design remains challenging without prior knowledge of the underlying physical phenomena. In this work a new data-based methodology is proposed involving the use of probabilistic, knowledge-guided artificial neural networks to jointly model the properties of the base components, the proportions in which they are mixed, and the processing conditions used during manufacture to predict properties of final products. The method proposed does not involve any assumptions in terms of ideal mixing rules of the base components, and allows for estimation of aleatoric uncertainty in the prediction. Additionally, an extension is presented that incorporates expert knowledge into the model by the implementation of monotonicity constraints between certain inputs and outputs. The methodology is illustrated with a case study involving the formulation of drug products using direct compression. The model is used to predict pharmaceutical tablets' quality attributes (mass variation, tensile strength, disintegration time, friability and ejection force), showing that the method is able to predict properties of the final product overcoming gaps currently present in previous modelling approaches., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier B.V. All rights reserved.)
Published: 2024
Full Text: View/download PDF

47. Advancing Precision Therapies in Brain Connectivity.

Author: Gómez F, Annen J, and Laureys S
Published: 2024
Full Text: View/download PDF

48. Bioaccumulation of organic and inorganic contaminants in biota: A long-term evaluation in the Belgian part of the North Sea.

Author: Vanavermaete D, De Witte B, Hostens K, Ruttens A, Waegeneers N, Cheyns K, Leus J, and De Baets B
Subjects: Animals, North Sea, Geologic Sediments chemistry, Crangonidae metabolism, Belgium, Metals metabolism, Metals analysis, Brachyura metabolism, Water Pollutants, Chemical analysis, Water Pollutants, Chemical metabolism, Polycyclic Aromatic Hydrocarbons analysis, Polycyclic Aromatic Hydrocarbons metabolism, Environmental Monitoring, Polychlorinated Biphenyls analysis, Polychlorinated Biphenyls metabolism, Bioaccumulation, Biota
Abstract: Determining the extent of pollution in the marine environment remains challenging. Polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), and metals can, during dredging operations, be transported from a port or harbour into the open sea, where they may exert a harmful effect on the marine ecosystem. To fully understand the impact of these chemicals, monitoring programs should not only focus on sediment concentrations, but should also take into account the bioaccumulated concentration in the tissue of multiple target organisms. In this study, the concentration of primary contaminants is determined in common starfish (Asterias rubens), flying crab (Liocarcinus holsatus), and brown shrimp (Crangon crangon) and the difference in the concentration of contamination between different dredge disposal sites at open sea is investigated. Different factors such as lipid weight, dry weight, grain size, and total organic carbon were measured and used to understand the difference between the observed bioaccumulation and the measured sediment concentrations. KEY MESSAGE: Different contaminants are detected in biota such as common starfish, flying crab and brown shrimp. These contaminants can be linked to dredging activities, with disposal sites associated with industrial ports showing higher contamination., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)
Published: 2024
Full Text: View/download PDF

49. Adjusting the timing of inseminations to the time lag on luteolysis alerts results in higher conception in dairy cattle.

Author: Meuwissen D, Gote MJ, Meyermans R, Janssens S, Adriaens I, and Aernouts B
Subjects: Animals, Cattle, Female, Pregnancy, Luteolysis, Retrospective Studies, Milk chemistry, Fertility, Time Factors, Fertilization, Insemination, Artificial veterinary, Progesterone
Abstract: Dairy cow fertility is a complex trait that depends on the cow's physiological status, the farm's environmental and management conditions, and their interactions. Already the slightest improvement in fertility can positively impact a farm's profitability and sustainability. In research, milk progesterone (P4) has often been used as an accurate and feasible way to identify a dairy cow's reproduction status. Moreover, in Europe and Canada, it has been used to improve fertility management on commercial farms as it allows to accurately identify reproduction issues, pregnancy, and the optimal insemination window. An on-farm P4 device (OPD) automatically samples, measures, and monitors the milk P4 concentration of individual cows. To this end, the P4 data are smoothed to be robust for measurement errors and outliers, and fixed thresholds are used to estimate the time of luteolysis preceding ovulation, thereby generating a luteolysis alert (LA). By smoothing the P4 data, the OPD introduces a time lag on the LA. Variation in this time lag is not considered in the estimation of the optimal insemination window that is advised to the farmer. Ignoring this variation might decrease the accuracy of the optimal insemination window and, therefore, decreases the likelihood of conception. We hypothesize that considering the length of the time lag and adapting the advice accordingly improves the conception rate. This observational retrospective study uses an extensive dataset from 17 commercial dairy farms that are equipped with an OPD. We estimated the time lag on the alerts and evaluated their relationship with the interval from LA to insemination for successful (n = 3,721) and unsuccessful inseminations (n = 3,896) separately. Results showed that the probability of conception increases when a longer LA time lag is compensated with a shorter interval from LA to insemination and vice versa. In addition, for successful inseminations, we found a clear negative relation between the time lag and the interval from LA to insemination and the interval was significantly shorter when the time lag of the LA was longer. This negative relation between time lag and interval from LA to insemination was less pronounced for unsuccessful inseminations. Additionally, we evaluated the conception rates for inseminations that are performed too early, in time, or too late with respect to the optimal insemination window advised by the OPD, in function of their associated time lags. We found that, for inseminations that were preceded by a short time lag (<8 h), the conception rate was 17.5 percentage points higher when cows were inseminated later than advised. Likewise, when inseminations were preceded by a long time lag (≥24 h), we found that the conception rate was 13 percentage points higher when cows were inseminated earlier than advised. Our results suggest that farmers using an OPD could potentially increase their conception success by compensating the variable time lag on the LA by adapting the interval from alert to insemination accordingly. This could be used to develop reproductive management strategies to improve reproductive performance on those farms, which can positively affect their sustainability., (© 2024, The Authors. Published by Elsevier Inc. on behalf of the American Dairy Science Association®. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).)
Published: 2024
Full Text: View/download PDF

50. Butterfly abundance changes in England are well associated with extreme climate events.

Author: Shan B, De Baets B, and Verhoest NEC
Subjects: England, Animals, Seasons, Ecosystem, Butterflies physiology, Climate Change
Abstract: Climate change exerts significant impacts on ecosystems through extreme climate events (ECEs), which are linked to various climate variables and can occur in different seasons. However, previous studies predominantly focus on a single type of ECE within specific seasons. We address this research gap by examining four typical types of ECEs: droughts, pluvials, heatwaves, and coldwaves, which are consistently defined and identified on a daily scale using a recently proposed statistical method. Butterflies in England serve as our study subject since their life stages occur throughout the year, and we had access to a 45-year dataset encompassing 57 butterfly species. First, we analyzed the correlation between abundance changes and the severity of ECEs, revealing varying sensitivity to ECEs across different life stages. Notably, abundance changes are negatively correlated with the severity of heatwaves in the hibernation, egg, and larval stages, as well as with the severity of pluvials in the larval, pupal, and adult stages. Second, we identified the most extreme climate events (MECEs) in England in the period from 1950 to 2020 and linked them with synchronized abundance changes for all species. MECEs in the sensitive stages of butterflies are frequently (for eight out of 11 MECEs) associated with synchronized decreases in their abundance. Our findings underscore the importance of considering diverse types of ECEs across all seasons to gain insights into their potential ecosystem effects. We draw attention to the fact that our analyses are primarily data-driven, with limited ecophysiological interpretation., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier B.V. All rights reserved.)
Published: 2024
Full Text: View/download PDF

Catalog

Books, media, physical & digital resources

See catalog results

Searchworks

Select search scope, currently: Articles Catalog books, media & more in Jio Institute collections Articles journal articles & other e-resources

Search

Search Constraints

Refine your results

Search Limiters

Topic

Publication Year Range

Language

Publication Type

Journal

Region

Database

Publisher

672 results on '"Department of Data Analysis"'

Search Results

Catalog

Select search scope, currently: Articles

Catalog

books, media & more in Jio Institute collections

Articles

journal articles & other e-resources