Search

Understanding the facial expressions of our interlocutor is important to enrich the communication and to give it a depth that goes beyond the explicitly expressed. In fact, studying one's facial expression gives insight into their hidden emotion state. However, even as humans, and despite our empathy and familiarity with the human emotional experience, we are only able to guess what the other might be feeling. In the fields of artificial intelligence and computer vision, Facial Emotion Recognition (FER) is a topic that is still in full growth mostly with the advancement of deep learning approaches and the improvement of data collection. The main purpose of this paper is to compare the performance of three state-of-the-art networks, each having their own approach to improve on FER tasks, on three FER datasets. The first and second sections respectively describe the three datasets and the three studied network architectures designed for an FER task. The experimental protocol, the results and their interpretation are outlined in the remaining sections.

Exploring the expansion history of the universe, understanding its evolutionary stages, and predicting its future evolution are important goals in astrophysics. Today, machine learning tools are used to help achieving these goals by analyzing transient sources, which are modeled as uncertain time series. Although black-box methods achieve appreciable performance, existing interpretable time series methods failed to obtain acceptable performance for this type of data. Furthermore, data uncertainty is rarely taken into account in these methods. In this work, we propose an uncertaintyaware subsequence based model which achieves a classification comparable to that of state-of-the-art methods. Unlike conformal learning which estimates model uncertainty on predictions, our method takes data uncertainty as additional input. Moreover, our approach is explainable-by-design, giving domain experts the ability to inspect the model and explain its predictions. The explainability of the proposed method has also the potential to inspire new developments in theoretical astrophysics modeling by suggesting important subsequences which depict details of light curve shapes. The dataset, the source code of our experiment, and the results are made available on a public repository.

Diagnosing erythema migrans (EM) skin lesion, the most common early symptom of Lyme disease using deep learning techniques can be effective to prevent long-term complications. Existing works on deep learning based EM recognition only utilizes lesion image due to the lack of a dataset of Lyme disease related images with associated patient data. Physicians rely on patient information about the background of the skin lesion to confirm their diagnosis. In order to assist the deep learning model with a probability score calculated from patient data, this study elicited opinion from fifteen doctors. For the elicitation process, a questionnaire with questions and possible answers related to EM was prepared. Doctors provided relative weights to different answers to the questions. We converted doctors evaluations to probability scores using Gaussian mixture based density estimation. For elicited probability model validation, we exploited formal concept analysis and decision tree. The elicited probability scores can be utilized to make image based deep learning Lyme disease pre-scanners robust.

Time series classification is a task that aims at classifying chronological data. It is used in a diverse range of domains such as meteorology, medicine and physics. In the last decade, many algorithms have been built to perform this task with very appreciable accuracy. However, applications where time series have uncertainty has been under-explored. Using uncertainty propagation techniques, we propose a new uncertain dissimilarity measure based on Euclidean distance. We then propose the uncertain shapelet transform algorithm for the classification of uncertain time series. The large experiments we conducted on state of the art datasets show the effectiveness of our contribution. The source code of our contribution and the datasets we used are all available on a public repository., Comment: 2020 International Conference on Data Mining Workshops, Sorrento, Italy, November 17-20, 2020

The gradual patterns that model the complex co-variations of attributes of the form "The more/less X, The more/less Y" play a crucial role in many real world applications where the amount of numerical data to manage is important, this is the biological data. Recently, these types of patterns have caught the attention of the data mining community, where several methods have been defined to automatically extract and manage these patterns from different data models. However, these methods are often faced the problem of managing the quantity of mined patterns, and in many practical applications, the calculation of all these patterns can prove to be intractable for the user-defined frequency threshold and the lack of focus leads to generating huge collections of patterns. Moreover another problem with the traditional approaches is that the concept of gradualness is defined just as an increase or a decrease. Indeed, a gradualness is considered as soon as the values of the attribute on both objects are different. As a result, numerous quantities of patterns extracted by traditional algorithms can be presented to the user although their gradualness is only a noise effect in the data. To address this issue, this paper suggests to introduce the gradualness thresholds from which to consider an increase or a decrease. In contrast to literature approaches, the proposed approach takes into account the distribution of attribute values, as well as the user's preferences on the gradualness threshold and makes it possible to extract gradual patterns on certain databases where literature approaches fail due to too large search space. Moreover, results from an experimental evaluation on real databases show that the proposed algorithm is scalable, efficient, and can eliminate numerous patterns that do not verify specific gradualness requirements to show a small set of patterns to the user., Comment: 24 pages, 12 figures

Time serie classification is used in a diverse range of domain such as meteorology, medicine and physics. It aims to classify chronological data. Many accurate approaches have been built during the last decade and shapelet transformation is one of them. However, none of these approaches does take data uncertainty into account. Using uncertainty propagation techiniques, we propose a new dissimilarity measure based on euclidean distance. We also show how to use this new measure to adapt shapelet transformation to uncertain time series classification. An experimental assessment of our contribution is done on some state of the art datasets., Comment: in French

In this paper, we propose a constraint-based modeling approach for the problem of discovering frequent gradual patterns in a numerical dataset. This SAT-based declarative approach offers an additional possibility to benefit from the recent progress in satisfiability testing and to exploit the efficiency of modern SAT solvers for enumerating all frequent gradual patterns in a numerical dataset. Our approach can easily be extended with extra constraints, such as temporal constraints in order to extract more specific patterns in a broad range of gradual patterns mining applications. We show the practical feasibility of our SAT model by running experiments on two real world datasets.

Clause Learning is one of the most important components of a conflict driven clause learning (CDCL) SAT solver that is effective on industrial instances. Since the number of learned clauses is proved to be exponential in the worse case, it is necessary to identify the most relevant clauses to maintain and delete the irrelevant ones. As reported in the literature, several learned clauses deletion strategies have been proposed. However the diversity in both the number of clauses to be removed at each step of reduction and the results obtained with each strategy creates confusion to determine which criterion is better. Thus, the problem to select which learned clauses are to be removed during the search step remains very challenging. In this paper, we propose a novel approach to identify the most relevant learned clauses without favoring or excluding any of the proposed measures, but by adopting the notion of dominance relationship among those measures. Our approach bypasses the problem of the diversity of results and reaches a compromise between the assessments of these measures. Furthermore, the proposed approach also avoids another non-trivial problem which is the amount of clauses to be deleted at each reduction of the learned clause database.

Recently, increasingly large amounts of data are generated from a variety of sources. Existing data processing technologies are not suitable to cope with the huge amounts of generated data. Yet, many research works focus on Big Data, a buzzword referring to the processing of massive volumes of (unstructured) data. Recently proposed frameworks for Big Data applications help to store, analyze and process the data. In this paper, we discuss the challenges of Big Data and we survey existing Big Data frameworks. We also present an experimental evaluation and a comparative study of the most popular Big Data frameworks. This survey is concluded with a presentation of best practices related to the use of the studied frameworks in several application domains such as machine learning, graph processing and real-world applications.

Big graph mining is an important research area and it has attracted considerable attention. It allows to process, analyze, and extract meaningful information from large amounts of graph data. Big graph mining has been highly motivated not only by the tremendously increasing size of graphs but also by its huge number of applications. Such applications include bioinformatics, chemoinformatics and social networks. One of the most challenging tasks in big graph mining is pattern mining in big graphs. This task consists on using data mining algorithms to discover interesting, unexpected and useful patterns in large amounts of graph data. It aims also to provide deeper understanding of graph data. In this context, several graph processing frameworks and scaling data mining/pattern mining techniques have been proposed to deal with very big graphs. This paper gives an overview of existing data mining and graph processing frameworks that deal with very big graphs. Then it presents a survey of current researches in the field of data mining / pattern mining in big graphs and discusses the main research issues related to this field. It also gives a categorization of both distributed data mining and machine learning techniques, graph processing frameworks and large scale pattern mining approaches., Comment: Submitted to Big Data Research, Elsevier

In Multiple Instance Learning (MIL) problem for sequence data, the instances inside the bags are sequences. In some real world applications such as bioinformatics, comparing a random couple of sequences makes no sense. In fact, each instance may have structural and/or functional relations with instances of other bags. Thus, the classification task should take into account this across bag relation. In this work, we present two novel MIL approaches for sequence data classification named ABClass and ABSim. ABClass extracts motifs from related instances and use them to encode sequences. A discriminative classifier is then applied to compute a partial classification result for each set of related sequences. ABSim uses a similarity measure to discriminate the related instances and to compute a scores matrix. For both approaches, an aggregation method is applied in order to generate the final classification result. We applied both approaches to solve the problem of bacterial Ionizing Radiation Resistance prediction. The experimental results of the presented approaches are satisfactory.

The choice of architecture of artificial neuron network (ANN) is still a challenging task that users face every time. It greatly affects the accuracy of the built network. In fact there is no optimal method that is applicable to various implementations at the same time. In this paper we propose a method to construct ANN based on clustering, that resolves the problems of random and ad hoc approaches for multilayer ANN architecture. Our method can be applied to regression problems. Experimental results obtained with different datasets, reveals the efficiency of our method.

This paper proposes a novel approach for storing and retrieving massive DNA sequences.. The method is based on a perceptual hash function, commonly used to determine the similarity between digital images, that we adapted for DNA sequences. Perceptual hash function presented here is based on a Discrete Cosine Transform Sign Only (DCT-SO). Each nucleotide is encoded as a fixed gray level intensity pixel and the hash is calculated from its significant frequency characteristics. This results to a drastic data reduction between the sequence and the perceptual hash. Unlike cryptographic hash functions, perceptual hashes are not affected by "avalanche effect" and thus can be compared. The similarity distance between two hashes is estimated with the Hamming Distance, which is used to retrieve DNA sequences. Experiments that we conducted show that our approach is relevant for storing massive DNA sequences, and retrieving them.

In the last few years, the amount of collected data, in various computer science applications, has grown considerably. These large volumes of data need to be analyzed in order to extract useful hidden knowledge. This work focuses on association rule extraction. This technique is one of the most popular in data mining. Nevertheless, the number of extracted association rules is often very high, and many of them are redundant. In this paper, we propose a new algorithm, called PRINCE. Its main feature is the construction of a partially ordered structure for extracting subsets of association rules, called generic bases. Without loss of information these subsets form representation of the whole association rule set. To reduce the cost of such a construction, the partially ordered structure is built thanks to the minimal generators associated to frequent closed patterns. The closed ones are simultaneously derived with generic bases thanks to a simple bottom-up traversal of the obtained structure. The experimentations we carried out in benchmark and "worst case" contexts showed the efficiency of the proposed algorithm, compared to algorithms like CLOSE, A-CLOSE and TITANIC., Comment: 50 pages, in French

With the emergence of graph databases, the task of frequent subgraph discovery has been extensively addressed. Although the proposed approaches in the literature have made this task feasible, the number of discovered frequent subgraphs is still very high to be efficiently used in any further exploration. Feature selection for graph data is a way to reduce the high number of frequent subgraphs based on exact or approximate structural similarity. However, current structural similarity strategies are not efficient enough in many real-world applications, besides, the combinatorial nature of graphs makes it computationally very costly. In order to select a smaller yet structurally irredundant set of subgraphs, we propose a novel approach that mines the top-k topological representative subgraphs among the frequent ones. Our approach allows detecting hidden structural similarities that existing approaches are unable to detect such as the density or the diameter of the subgraph. In addition, it can be easily extended using any user defined structural or topological attributes depending on the sought properties. Empirical studies on real and synthetic graph datasets show that our approach is fast and scalable.

The increasing growth of databases raises an urgent need for more accurate methods to better understand the stored data. In this scope, association rules were extensively used for the analysis and the comprehension of huge amounts of data. However, the number of generated rules is too large to be efficiently analyzed and explored in any further process. Association rules selection is a classical topic to address this issue, yet, new innovated approaches are required in order to provide help to decision makers. Hence, many interesting- ness measures have been defined to statistically evaluate and filter the association rules. However, these measures present two major problems. On the one hand, they do not allow eliminating irrelevant rules, on the other hand, their abun- dance leads to the heterogeneity of the evaluation results which leads to confusion in decision making. In this paper, we propose a two-winged approach to select statistically in- teresting and semantically incomparable rules. Our statis- tical selection helps discovering interesting association rules without favoring or excluding any measure. The semantic comparability helps to decide if the considered association rules are semantically related i.e comparable. The outcomes of our experiments on real datasets show promising results in terms of reduction in the number of rules.

In folksonomies, users use to share objects (movies, books, bookmarks, etc.) by annotating them with a set of tags of their own choice. With the rise of the Web 2.0 age, users become the core of the system since they are both the contributors and the creators of the information. Yet, each user has its own profile and its own ideas making thereby the strength as well as the weakness of folksonomies. Indeed, it would be helpful to take account of users' profile when suggesting a list of tags and resources or even a list of friends, in order to make a personal recommandation, instead of suggesting the more used tags and resources in the folksonomy. In this paper, we consider users' profile as a new dimension of a folksonomy classically composed of three dimensions and we propose an approach to group users with equivalent profiles and equivalent interests as quadratic concepts. Then, we use such structures to propose our personalized recommendation system of users, tags and resources according to each user's profile. Carried out experiments on two real-world datasets, i.e., MovieLens and BookCrossing highlight encouraging results in terms of precision as well as a good social evaluation., Comment: 7 pages

One of the most powerful techniques to study protein structures is to look for recurrent fragments (also called substructures or spatial motifs), then use them as patterns to characterize the proteins under study. An emergent trend consists in parsing proteins three-dimensional (3D) structures into graphs of amino acids. Hence, the search of recurrent spatial motifs is formulated as a process of frequent subgraph discovery where each subgraph represents a spatial motif. In this scope, several efficient approaches for frequent subgraph discovery have been proposed in the literature. However, the set of discovered frequent subgraphs is too large to be efficiently analyzed and explored in any further process. In this paper, we propose a novel pattern selection approach that shrinks the large number of discovered frequent subgraphs by selecting the representative ones. Existing pattern selection approaches do not exploit the domain knowledge. Yet, in our approach we incorporate the evolutionary information of amino acids defined in the substitution matrices in order to select the representative subgraphs. We show the effectiveness of our approach on a number of real datasets. The results issued from our experiments show that our approach is able to considerably decrease the number of motifs while enhancing their interestingness.

Folksonomy mining is grasping the interest of web 2.0 community since it represents the core data of social resource sharing systems. However, a scrutiny of the related works interested in mining folksonomies unveils that the time stamp dimension has not been considered. For example, the wealthy number of works dedicated to mining tri-concepts from folksonomies did not take into account time dimension. In this paper, we will consider a folksonomy commonly composed of triples and we shall consider the time as a new dimension. We motivate our approach by highlighting the battery of potential applications. Then, we present the foundations for mining quadri-concepts, provide a formal definition of the problem and introduce a new efficient algorithm, called QUADRICONS for its solution to allow for mining folksonomies in time, i.e., d-folksonomies. We also introduce a new closure operator that splits the induced search space into equivalence classes whose smallest elements are the quadri-minimal generators. Carried out experiments on large-scale real-world datasets highlight good performances of our algorithm.

Recently, graph mining approaches have become very popular, especially in domains such as bioinformatics, chemoinformatics and social networks. In this scope, one of the most challenging tasks is frequent subgraph discovery. This task has been motivated by the tremendously increasing size of existing graph databases. Since then, an important problem of designing efficient and scaling approaches for frequent subgraph discovery in large clusters, has taken place. However, failures are a norm rather than being an exception in large clusters. In this context, the MapReduce framework was designed so that node failures are automatically handled by the framework. In this paper, we propose a large-scale and fault-tolerant approach of subgraph mining by means of a density-based partitioning technique, using MapReduce. Our partitioning aims to balance computation load on a collection of machines. We experimentally show that our approach decreases significantly the execution time and scales the subgraph discovery process to large graph databases., Comment: The paper is under reviewing and we want to cancel the submission. Thank you for your understanding

Finding interesting association rules is an important and active research field in data mining. The algorithms of the Apriori family are based on two rule extraction measures, support and confidence. Although these two measures have the virtue of being algorithmically fast, they generate a prohibitive number of rules most of which are redundant and irrelevant. It is therefore necessary to use further measures which filter uninteresting rules. Many synthesis studies were then realized on the interestingness measures according to several points of view. Different reported studies have been carried out to identify "good" properties of rule extraction measures and these properties have been assessed on 61 measures. The purpose of this paper is twofold. First to extend the number of the measures and properties to be studied, in addition to the formalization of the properties proposed in the literature. Second, in the light of this formal study, to categorize the studied measures. This paper leads then to identify categories of measures in order to help the users to efficiently select an appropriate measure by choosing one or more measure(s) during the knowledge extraction process. The properties evaluation on the 61 measures has enabled us to identify 7 classes of measures, classes that we obtained using two different clustering techniques., Comment: 34 pages, 4 figures

Feature extraction is an unavoidable task, especially in the critical step of preprocessing biological sequences. This step consists for example in transforming the biological sequences into vectors of motifs where each motif is a subsequence that can be seen as a property (or attribute) characterizing the sequence. Hence, we obtain an object-property table where objects are sequences and properties are motifs extracted from sequences. This output can be used to apply standard machine learning tools to perform data mining tasks such as classification. Several previous works have described feature extraction methods for bio-sequence classification, but none of them discussed the robustness of these methods when perturbing the input data. In this work, we introduce the notion of stability of the generated motifs in order to study the robustness of motif extraction methods. We express this robustness in terms of the ability of the method to reveal any change occurring in the input data and also its ability to target the interesting motifs. We use these criteria to evaluate and experimentally compare four existing extraction methods for biological sequences., Comment: The paper has been accepted by the ACM Conference on Bioinformatics, Computational Biology and Biomedicine (ACM BCB) 2012. We want to cancel the submission because of the double entries of the paper in DBLP. Thank you for your understanding

Formal Concept Analysis "FCA" is a data analysis method which enables to discover hidden knowledge existing in data. A kind of hidden knowledge extracted from data is association rules. Different quality measures were reported in the literature to extract only relevant association rules. Given a dataset, the choice of a good quality measure remains a challenging task for a user. Given a quality measures evaluation matrix according to semantic properties, this paper describes how FCA can highlight quality measures with similar behavior in order to help the user during his choice. The aim of this article is the discovery of Interestingness Measures "IM" clusters, able to validate those found due to the hierarchical and partitioning clustering methods "AHC" and "k-means". Then, based on the theoretical study of sixty one interestingness measures according to nineteen properties, proposed in a recent study, "FCA" describes several groups of measures., Comment: 13 pages, 2 figures