Search

In today's world where information is a valuable and vital resource, Information Management emerges as a fundamental discipline to deal with the growing volume of data and transform it into strategic knowledge. This article seeks to present the development of Information Management from the emergence of Information Science and its subareas and its contribution to the management of organizations. Thus, the work unfolds into four main sections: (1) discussion of the framework of the emergence of Information Science and its main milestones over the years; (2) different concepts and characteristics of IC and the emergence of different branched subareas of IC; (3) presentation of the definitions and characteristics of the Information Management subarea; (4) exposition of the various advantages resulting from the application of Information Management in organizations. From the research it was possible to identify that the main advantages arising from information management contribute especially to a better understanding of the different organizational areas and help in decision making. [ABSTRACT FROM AUTHOR]

This article addresses the intersection between Hermeneutics and Information Organization and Representation, emphasizing the importance of transversality and verticality in the context of Information Science. Hermeneutics, as an interpretative approach, is essential to comprehend the complexity of information, considering cultural, historical, and social contexts. Simultaneously, Information Organization and Representation aim to structure and give meaning to data, facilitating access and understanding. Transversality refers to the integration of these elements horizontally, while verticality involves depth in the analysis and interpretation of information. The combination of these approaches enriches research in Information Science, providing a more comprehensive and contextualized understanding of the crucial role of information in contemporary society. [ABSTRACT FROM AUTHOR]

Purpose: This paper analyses the research in Library and Information Science (LIS) and reports on (1) the status of LIS research in 2015 and (2) on the evolution of LIS research longitudinally from 1965 to 2015. Design/methodology/approach: The study employs a quantitative intellectual content analysis of articles published in 30+ scholarly LIS journals, following the design by Tuomaala et al. (2014). In the content analysis, we classify articles along eight dimensions covering topical content and methodology. Findings: The topical findings indicate that the earlier strong LIS emphasis on L&I services has declined notably, while scientific and professional communication has become the most popular topic. Information storage and retrieval has given up its earlier strong position towards the end of the years analyzed. Individuals are increasingly the units of observation. End-user's and developer's viewpoints have strengthened at the cost of intermediaries' viewpoint. LIS research is methodologically increasingly scattered since survey, scientometric methods, experiment, case studies and qualitative studies have all gained in popularity. Consequently, LIS may have become more versatile in the analysis of its research objects during the years analyzed. Originality/value: Among quantitative intellectual content analyses of LIS research, the study is unique in its scope: length of analysis period (50 years), width (8 dimensions covering topical content and methodology) and depth (the annual batch of 30+ scholarly journals). [ABSTRACT FROM AUTHOR]

Users increasingly acquire, share, and create knowledge in online knowledge communities, making them massive dynamic knowledge repositories that spark inspiration. While online knowledge communities provide powerful searching tools, they ignore the potential of serendipity in fostering knowledge acquisition. Against this backdrop, this research investigates whether serendipity can be stimulated by design features of communities. Specifically, we examine whether different tag presentation formats may promote serendipity. Two hundred seven participants were randomly assigned to our experimental website that displays one of three tag formats. Results show that users experienced serendipity more frequently while using tag trees than tag clouds, followed by tag lists. Moreover, tag formats moderate how openness to experience affects serendipity. Although openness did not influence serendipity across tag formats, further analysis shows that it significantly decreases serendipity for tag lists, but significantly increases serendipity for tag clouds and trees. Theoretically, these results provide an in‐depth understanding of serendipity that is contingent on the interaction between community design features and personality (e.g., openness to experience). Practically, these findings demonstrate how interface features (e.g., tag presentation formats) facilitate serendipity, thus informing better design of online knowledge communities to improve the efficiency of knowledge acquisition. [ABSTRACT FROM AUTHOR]

The purpose of this article is to delineate the process of evolution of knowledge organization systems (KOSs) through identification of principles of unity such as internal and external unity in organizing the structure of KOSs to achieve content storage and retrieval purposes and to explain a novel method used in ranking of KOSs by proposing the principle of rank unity. Different types of KOSs which are addressed in this article include dictionaries, Roget's thesaurus, thesauri, micro, macro, and meta-thesaurus, ontologies, and lower, middle, and upper-level ontologies. This article relied on dialectic models to clarify the ideas in Kant's knowledge theory. This is done by identifying logical relationships between categories (i.e., Thesis, antithesis, and synthesis) in the creation of data, information, and knowledge in the human mind. The Analysis has adapted a historical methodology, more specifically a documentary method, as its reasoning process to propose a conceptual model for ranking KOSs. The study endeavors to explain the main elements of data, information, and knowledge along with engineering mechanisms such as data, information, and knowledge engineering in developing the structure of KOSs and also aims to clarify their influence on content storage and retrieval performance. KOSs have followed related principles of order to achieve an internal order, which could be examined by analyzing the principle of internal unity in knowledge organizations. The principle of external unity leads us to the necessity of compatibility and interoperability between different types of KOSs to achieve semantic harmonization in increasing the performance of content storage and retrieval. Upon introduction of the principle of rank unity, a ranking method of KOSs utilizing cognition states as criteria could be considered to determine the position of each knowledge organization with respect to others. The related criteria of the principle of rank unity- cognition states- are derived from Immanuel Kant's epistemology. The research results showed that KOSs, while having defined positions in cognition states, specific principles of order, related operational mechanisms, and related principles of unity in achieving their specific purposes, have benefited from the developmental experiences of previous KOSs, and further, their developmental processes owe to the experiences and methods of their previous generations. [ABSTRACT FROM AUTHOR]

Copyright of Revista Ibero-Americana de Ciência da Informação is the property of Revista Ibero-Americana de Ciencia da Informacao and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Revista de Tecnologia Aplicada is the property of Faculdade Campo Limpo Paulista and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

International comparative research in the library and information science (LIS) field examines the processes and phenomena related to libraries and other information organizations and their users. with a focus on differences and similarities across countries or cultures. International research is challenging due to language barriers, ethical concerns, and the legacy of the colonial research model. This paper presents an international research project undertaken by members of the International Federation of Library Associations (IFLA) Library Theory and Research (LTR) Section which investigated the approaches to teaching research methods in LIS programs worldwide. The paper focuses on the project's research design, on the research ethical issues and on the collection of multilingual data. It discusses the inherent challenges in conducting international research and outlines the approach to increasing the geographic and linguistic diversity of study respondents. The LTR research team adopted several strategies to recruit participants from multiple countries and collect data in three languages. The recruitment announcements were distributed throughout international and regional mailing lists in multiple languages. The survey instrument was translated from English to Spanish and French, and the interviews were conducted in English and Spanish. The authors also discuss the methodological advantages of mixed-methods design and the benefits and limitations of using surveys and interviews in international research. [ABSTRACT FROM AUTHOR]

Bibliometrics and content analysis approaches were adopted in this study to review the papers issued in Journal of Library and Information Sciences from 1999 to 2020 to draw the landscape of library and information science studies, including research population, research subjects, research approaches and theories used. The results showed the growth of the research population, but the high percentage of one-time authors who lacked sustainable research efforts. Classical LIS topics, such as information behavior and information organization with mixed and newly developed approaches were the center of research. Research evaluation by bibliometrics approach expanded the scope of LIS studies. The presentation of theory linkage appeared limited, but increased in later published papers. [ABSTRACT FROM AUTHOR]

This study investigates organizational patterns from 12,781 subject guides at 114 academic institutions, comparing practices evident on LibGuides with stated best practices from the literature of library and information science. Data from subject guide fields such as titles, tabs, and boxes were collected systematically from disciplinary guides at institutions in the United States that are described in Carnegie Classifications as "R1: Doctoral Universities – Very high research activity," and then analyzed according to two general methods. First, descriptive statistics were generated for quantitative aspects of the LibGuides data (e.g., the average number of boxes used per guide), to map both normal and atypical organizational practices. Second, word and phrase frequencies from text fields were compiled to explore differences in how fields such as guide titles and descriptions are commonly utilized. The findings suggest that the average range of guides tends to follow best practices, but that in the use of certain guide elements, especially tabs, most guides do not follow recommended guidelines. In addition, the data identify a number of guides that represent extreme outliers to typical practices: a single guide contained 144 tabs, for example, while another subject guide landing page was found to include 2,033 links. [ABSTRACT FROM AUTHOR]

Copyright of Turismo: Visão e Ação is the property of Turismo-Visao E Acao and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Theoretical foundations and philosophical basis of library classification systems have always been important for information science professionals. This research studies the philosophical foundations and logical basis of the classification systems and structure of human knowledge and studies the logical basis of library classifications Scheme. Challenges and problems of traditional ideas about knowledge organization and its hierarchical structure have been studied and rhizomatic thought has been proposed as an alternative model for the tree and hierarchical structure especially for the organization of digital and hypertext information on the web. In Western thought, the tree structure based on Aristotelian logic has been the dominant model of classification and due to its problems, the rhizomatic structure is suggested. Information structure in digital space such as rhizomes moves in many directions and creates a multiplicity of connections and relations. The growth and emergence of interdisciplinary studies can be seen as evidence. Traditional, tree-based and hierarchical discourse of information organization isn’t enabled to understand the nature and characteristics of the digital information space and so to organize it. Necessarily to a paradigm shift, rhizomatic thinking can be a good alternative to traditional information organizing thinking. [ABSTRACT FROM AUTHOR]

This article offers an analysis of the process of error recovery as observed in the development and use of collections databases in a university museum. It presents results from a longitudinal case study of the development of collaborative systems and practices designed to reduce the number of errors found in the museum's databases as museum employees packed and moved their collections over a period of five years. Drawing on a specific set of eighteen months' worth of records documenting error rates and error management techniques, this article identifies three factors that influenced the ability of museum staff members to recover from errors found in their collections databases. The article concludes by examining the potential impact of these factors on the design of future collections databases in order to shed light on the wider issue of error recovery procedures in all information organizations. [ABSTRACT FROM AUTHOR]

The International Conference on Technology and Information Organization is an initiative of the "Observatory of the Labor Market in Information and Documentation" research group (OMTID -- CNPq) of the School of Communications and Arts of the University of São Paulo. The TOI take place in academic and scientific collaboration with the CONTECSI FEA-USP and CIC.DIGITAL -- Porto University, and the Laboratório de Tecnologias Intelectuais LTi from UFPB, bringing together Information Science researchers, students and professionals - Librarianship, Archival Science and Museology, with the goal of promoting reflection and dialogue about relevant topics, as well as contributing to the integration of the academic and the professional environments, strengthening the interest in research, and sharing knowledge about the most innovative practices in this area. [ABSTRACT FROM AUTHOR]

Copyright of Museologia e Patrimônio is the property of Museologia e Patrimonio and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This document is a correction notice for an article titled "Accelerated preprocessing of large numbers of brain images by parallel computing on supercomputers" published in Scientific Reports. The correction addresses an error in the Acknowledgements section of the original article. The corrected version acknowledges the use of computational resources from the supercomputer "Fugaku" provided by the RIKEN Center for Computational Science, as well as support from various research organizations and grants in Japan. The correction notice is authored by Takehiro Jimbo, Hidetoshi Matsuo, Yuya Imoto, Takumi Sodemura, Makoto Nishimori, Yoshinari Fukui, Takuya Hayashi, Tomoyuki Furuyashiki, and Ryoichi Yokoyama. [Extracted from the article]

The article discusses facet theory as a way of developing theoretical constructs in management theory. He notes that facet theory derives from set theory, and that it provides a useful answer to the question of what properties a definition must have if it is to be useful to the scientific community. He observes that the value of a definition is in its ability to distinguish between bad and good instances of a behavioral construct. An example he cites of using facet theory is that of sentence mapping.

It investigates how the Information Science in Brazil establishes its interdisciplinary dynamics with the French Discourse Analysis, which has unfolded in different theoretical and methodological orders. In addition, it analyzes, from a diachronic perspective, the epistemological interactions between the French aspect of Discourse Analysis and Information Science, especially with regard to the Organization of Information and Knowledge. Through a literature review, it shows the main historical contributions of discourse theory to the informational field in the Brazilian context. Among the theoretical contributions of Discourse Analysis is the The article explores the dialogical relationship that can be established between the French theory of discourse analysis in the Information and Knowledge Organization in the Brazilian context. In the broad sense, the reflective path comprises a possible gnosiology present in the Science of Information and the discursive studies. The focus of this analytical study is on the interdisciplinary issues, as well as on the epistemological relationships, that this dialogue presents for Documentary Analysis. The objective is to establish the interdisciplinary relations of the two areas of study in the Brazilian context, having as a methodological path for the understanding of this diachronic perspective the critical epistemology. Three lines of discourse analysis were chosen for the purpose of the dialogical relationship: the line of Michel Pêcheux, the line of Sociolinguistics and the line of Michel Foucault. The first and third are present in works carried out in the field of Information Science, especially in the convergences with Documentary Analysis, having as confluence information as discourse, subject, language and pragmatics. Possible paths point to a scenario affected by power disputes. Thus, this scenario produces meaning effects and needs to be analyzed by a reading gesture, taking into account the historical issues of time and space, because depending on the cultural and temporal moment, information and discourse can produce different effects of meaning. [ABSTRACT FROM AUTHOR]

One critical deficiency of traditional online kernel learning methods is their unbounded and growing number of support vectors in the online learning process, making them inefficient and non-scalable for large-scale applications. Recent studies on scalable online kernel learning have attempted to overcome this shortcoming, e.g., by imposing a constant budget on the number of support vectors. Although they attempt to bound the number of support vectors at each online learning iteration, most of them fail to bound the number of support vectors for the final output hypothesis, which is often obtained by averaging the series of hypotheses over all the iterations. In this article, we propose a novel framework for bounded online kernel methods, named “Sparse Passive-Aggressive (SPA)” learning, which is able to yield a final output kernel-based hypothesis with a bounded number of support vectors. Unlike the common budget maintenance strategy used by many existing budget online kernel learning approaches, the idea of our approach is to attain the bounded number of support vectors using an efficient stochastic sampling strategy that samples an incoming training example as a new support vector with a probability proportional to its loss suffered. We theoretically prove that SPA achieves an optimal mistake bound in expectation, and we empirically show that it outperforms various budget online kernel learning algorithms. Finally, in addition to general online kernel learning tasks, we also apply SPA to derive bounded online multiple-kernel learning algorithms, which can significantly improve the scalability of traditional Online Multiple-Kernel Classification (OMKC) algorithms while achieving satisfactory learning accuracy as compared with the existing unbounded OMKC algorithms. [ABSTRACT FROM AUTHOR]

Purpose - The purpose of this chapter is to provide a conceptual exploration of cataloging and metadata education. Historically and currently, cataloging and metadata are an essential part of master's-level library and information science (LIS) education. Design/Methodology/Approach - We review LIS literature and provide evidence from their own experiences to support their argument. Findings - Cataloging education, far from going the way of the dodo, is still a very important part of LIS education. Even though general information organization courses are still required by most LIS programs, cataloging and metadata courses that include a balance of theory and practice are often buried as electives within LIS school curricula. Information organization principles and techniques (both theory and practice) are highly relevant in today's information environment. Originality/Value - This chapter was written by four cataloging educators, who have extensive cataloging knowledge and experience and who have seen firsthand the benefits of cataloging education for all LIS students. As library professionals adapt, and given the increasing focus on users and their needs, the relevance and necessity of a robust understanding of cataloging and metadata creation principles is key going forward. [ABSTRACT FROM AUTHOR]

Copyright of Informacao & Sociedade: Estudos is the property of Universidade Federal de Campina Grande, Centro de Humanidades and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Shelf arrangement, from a bibliographic perspective, constitutes a reading proposal of the collection to the users as well as a resource for management and access to the documents. However, the centrality of the call number testifies the near forgetfulness of the different proposals that came before it and the role of the collection of documents and the target audience in the elaboration of the organization, in addition to the overlapping of the bibliographic classification to shelf arrangement. This work is justified by the need to restore shelf arrangement, seeking to understand its fundamental aspects from the literature in which the activity was systematized. Thus, this paper aims at contributing to reorient the shelf arrangement as an activity of information organization, exploring its conformation in the Francophone literature, from the mid-nineteenth century up to the 1930s. As for the methodology, this is an exploratory research made possible through the historicalconceptual investigation of shelf arrangement found in the Francophone manuals of librarianship of that period. This study concludes that the activity was placed by that line since the nineteenth century, when its own terminology was developed under the consideration of the intervention of the contexts, using methods and guided by the diversity of proposals. [ABSTRACT FROM AUTHOR]

Hyperspectral images comprise hundreds of narrow contiguous wavelength bands which include wealth spectral information, and a great potential of Light detection and ranging (LIDAR) data lies in its benefits of height measurements, which can be used as complementary information for the classification of hyperspectral data. In this paper, a feature-fusion strategy of hyperspectral and LIDAR data is taken into account in order to develop a new classification framework for the accurate analysis of a surveyed area. The proposed approach employs extinction profiles (EPs) extracted with extinction filters computed on both hyperspectral and LIDAR images, leading to a fusion of the spectral, spatial, and elevation features. Experimental results obtained by using a real hyperspectral image along with LIDAR-derived digital surface model (DSM) collected over the University of Houston campus and its neighboring urban area demonstrate the effectiveness of the proposed framework. [ABSTRACT FROM AUTHOR]

A class of graphs is called monotone if it is closed under deletion of vertices and edges. Any such class may be defined in terms of forbidden subgraphs. The chromatic index of a graph is the smallest number of colors required for its edge-coloring such that any two adjacent edges have different colors. We obtain a complete classification of the complexity of the chromatic index problem for all monotone classes defined in terms of forbidden subgraphs having at most 6 edges or at most 7 vertices. [ABSTRACT FROM AUTHOR]

We analyze the detection and classification of singularities of functions f = χ S , where S ⊂ R d and d = 2 , 3 . It will be shown how the set ∂ S can be extracted by a continuous shearlet transform associated with compactly supported shearlets. Furthermore, if ∂ S is a d − 1 dimensional piecewise smooth manifold with d = 2 or 3, we will classify smooth and non-smooth components of ∂ S . This improves previous results given for shearlet systems with a certain band-limited generator, since the estimates we derive are uniform. Moreover, we will show that our bounds are optimal. Along the way, we also obtain novel results on the characterization of wavefront sets in 3 dimensions by compactly supported shearlets. Finally, geometric properties of ∂ S such as curvature are described in terms of the continuous shearlet transform of f . [ABSTRACT FROM AUTHOR]

Let Sn be the symmetric group of all permutations of {1, ..., n}. We show that the number of distinct Wilf classes of subsets of exactly 8 and 9 four-letter patterns is 19002 and 16293, respectively. [ABSTRACT FROM AUTHOR]

Nowadays, typical active contour models are widely applied in image segmentation. However, they perform badly on real images with inhomogeneous subregions. In order to overcome the drawback, this paper proposes an edge-preserving smoothing image segmentation algorithm. At first, this paper analyzes the edge-preserving smoothing conditions for image segmentation and constructs an edge-preserving smoothing model inspired by total variation. The proposed model has the ability to smooth inhomogeneous subregions and preserve edges. Then, a kind of clustering algorithm, which reasonably trades off edge-preserving and subregion-smoothing according to the local information, is employed to learn the edge-preserving parameter adaptively. At last, according to the confidence level of segmentation subregions, this paper constructs a smoothing convergence condition to avoid oversmoothing. Experiments indicate that the proposed algorithm has superior performance in precision, recall, and F-measure compared with other segmentation algorithms, and it is insensitive to noise and inhomogeneous-regions. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI. [DOI: 10.1117/1.JEI.25.5.053022] [ABSTRACT FROM AUTHOR]

In this paper, a comparative analysis of different texture features based on local operator has been produced for the determination of mammographic masses as benign or malignant. Local Binary Pattern (LBP), LBP Variance (LBPV), and Completed LBP (CLBP) descriptors are extracted to evaluate their potential for mass classification in a Computer-Aided Diagnosis (CAD) system. An Az value of 0 . 97 ± 0 . 02 and an accuracy of 92 . 25 ± 0 . 01% have been achieved, while experimenting on 200 mass cases from the DDSM database, by selecting the optimal set of features employing stepwise logistic regression method, followed by classification via Fisher Linear Discriminant Analysis (FLDA) using 10-fold cross validation. [ABSTRACT FROM AUTHOR]

Libraries have long struggled with the question of how to best classify and enable access to videorecordings. While giving a presentation at a state library conference, the authors observed from comments during the question and answer portion of the presentation that the libraries represented in the audience use a variety of video classification and organization practices. To better understand how local practices and librarians' attitudes regarding the efficacy of these practices vary, we conducted a survey soliciting responses from librarians representing a broad array of library types. [ABSTRACT FROM AUTHOR]

This paper calls for a review of curriculum in the discipline of information studies to respond to the recent changes brought by digital information environment, social media, and multidisciplinary initiatives. This is an opinion paper based mainly on author's extensive teaching experience and research in the area of information/knowledge organization. Relevant literature was reviewed to develop theoretical context and support assertions made and suggestions put forward by the author. This paper makes a case for fundamental changes in curriculum in the area of information organization to expand the scope of core knowledge in this important area. In author's opinion, reconceptualization of contents has become necessary to respond to changes taking place in the information field in the digital environment. Author suggests that new topics are placed in different courses in the information studies programs (introductory/foundation; core/required; and electives/specialized courses) for a balanced approach. Discussion on possible changes will be helpful in curriculum design and teaching of information/knowledge organization courses in information studies ram to make these courses more useful for professionals aspiring to work in modern information environments. [ABSTRACT FROM AUTHOR]

Innovative data mining and knowledge discovery approaches that take advantage of geospatial analysis can be useful for analysis of motor vehicle crashes on regional highway corridors. This study presents an Ontology-based Classification and Regression Tree ( OCART) approach to induct crash rules on regional highway corridors by adding ontological reasoning to Classification and Regression Tree ( CART) decision rules. It develops an ontology-driven geospatial framework to predict the motor vehicles crash severity through the proposed OCART approach. A system prototype has been developed and implemented on a regional highway corridor in order to illustrate and evaluate the proposed method. The results demonstrate the application of ontological reasoning and spatial computation significantly improves the performance of CART crash rules. The proposed method reveals new relationships among crash severities and contributing factors that have been kept implicit in CART decision rules. WIREs Data Mining Knowl Discov 2015, 5:181-194. doi: 10.1002/widm.1152 For further resources related to this article, please visit the . [ABSTRACT FROM AUTHOR]

In research on multiattribute decisions, information is typically preorganized in a well-structured manner (e.g., in attributes-by-options matrices). Participants can therefore conveniently identify the information needed for the decision strategy they are using. However, in everyday decision situations, we often face information that is not well-structured; that is, we not only have to search for, but we also need to organize the information. This latter aspect--subjective information organization--has so far largely been neglected in decision research. The few exceptions used crude experimental manipulations, and the assessment of subjective organization suffered from laborious methodology and a lack of objectiveness. We introduce a new task format to overcome these methodological issues, and we provide an organization index (OI) to assess subjective organization of information objectively and automatically. The OI makes it possible to assess information organization on the same scale as the strategy index (SI) typically used for assessing information search behavior. A simulation study shows that the OI has a similar distribution as the SI but that the two indices are a priori largely independent. In a validation experiment with instructed strategy use, we demonstrate the usefulness of the task to trace decision processes in multicue inference situations. [ABSTRACT FROM AUTHOR]

This article presents a new semidistance for functional observations that generalizes the Mahalanobis distance for multivariate datasets. The main characteristics of the functional Mahalanobis semidistance are shown. To illustrate the applicability of this measure of proximity between functional observations, new versions of several well-known functional classification procedures are developed using the functional Mahalanobis semidistance. A Monte Carlo study and the analysis of two real examples indicate that the classification methods used in conjunction with the functional Mahalanobis semidistance give better results than other well-known functional classification procedures. This article has supplementary material online. [ABSTRACT FROM PUBLISHER]

In this work, we propose a method for the classification of animal in images. Initially, a graph cut based method is used to perform segmentation in order to eliminate the background from the given image. The segmented animal images are partitioned in to number of blocks and then the color texture moments are extracted from different blocks. Probabilistic neural network and K-nearest neighbors are considered here for classification. To corroborate the efficacy of the proposed method, an experiment was conducted on our own data set of 25 classes of animals, which consisted of 4000 sample images. The experiment was conducted by picking images randomly from the database to study the effect of classification accuracy, and the results show that the K-nearest neighbors classifier achieves good performance. [ABSTRACT FROM AUTHOR]

We investigate how to iteratively and mutually boost object classification and detection performance by taking the outputs from one task as the context of the other one. While context models have been quite popular, previous works mainly concentrate onco-occurrence relationship within classes and few of them focus on contextualization from a top-down perspective, i.e. high-level task context. In this paper, our system adopts a new method for adaptive context modeling and iterative boosting. First, the contextualized support vector machine (Context-SVM) is proposed, where the context takes the role of dynamically adjusting the classification score based on the sample ambiguity, and thus the context-adaptive classifier is achieved. Then, an iterative training procedure is presented. In each step, Context-SVM, associated with the output context from one task (object classification or detection), is instantiated to boost the performance for the other task, whose augmented outputs are then further used to improve the former task by Context-SVM. The proposed solution is evaluated on the object classification and detection tasks of PASCAL Visual Object Classes Challenge (VOC) 2007, 2010 and SUN09 data sets, and achieves the state-of-the-art performance. [ABSTRACT FROM AUTHOR]

We propose a novel hybrid loss for multiclass and structured prediction problems that is a convex combination of a log loss for Conditional Random Fields (CRFs) and a multiclass hinge loss for Support Vector Machines (SVMs). We provide a sufficient condition for when the hybrid loss is Fisher consistent for classification. This condition depends on a measure of dominance between labels—specifically, the gap between the probabilities of the best label and the second best label. We also prove Fisher consistency is necessary for parametric consistency when learning models such as CRFs. We demonstrate empirically that the hybrid loss typically performs least as well as—and often better than—both of its constituent losses on a variety of tasks, such as human action recognition. In doing so we also provide an empirical comparison of the efficacy of probabilistic and margin based approaches to multiclass and structured prediction. [ABSTRACT FROM AUTHOR]