Web service discovery in dynamic and large-scale environments poses significant challenges due to the continuous evolution of services and the diversity of user needs. This paper proposes an incremental, community-driven approach for context-aware service discovery. By modeling user-service interactions and semantic descriptions (SAWSDL) within a heterogeneous graph, the system detects and maintains functional communities through incremental label propagation. Each service and user is dynamically labeled to reflect evolving usage patterns and semantic domains. When a service request is issued, discovery is performed within the most relevant community, and candidate services are ranked using both semantic similarity and a novel Community Impact Index (IIC), which reflects the expertise of users within the target community.

Considerable and high-quality scientific research papers play a crucial role in acquiring knowledge across various fields. Extracting information from the full text of these papers presents a challenging task. Named Entity Recognition (NER) stands out as a key step in information extraction. This recognition is indispensable for many tasks and applications requiring direct access to relevant information, such as knowledge discovery and document retrieval. However, effectively processing important contextual information remains a challenge in the literature. In our work, we introduce a supervised NER method named N-AMES (Named entity recognition using contextual Attention on Masked Entities and Sections). Leveraging the attention mechanism of the BERT transformer, our method effectively processes both global and local entity context information. By incorporating section titles into a sequence of text chunks and training the model on masked entities, our approach achieves remarkable performance. Specifically, our model achieves an F1-measure of 74.10% and 72.30% in partial matching evaluation, outperforming state-of-the-art models on two distinct full-text research paper datasets: SciREX for machine learning and CRAFT for biomedical entities.

Influence maximization is an important research area in social network analysis where researchers are concerned with detecting influential nodes . The detection of influential nodes is of great interest in several disciplines including computer science, opinion propagation, political movements , or economics, where systems are often modeled as graphs. The Influence Maximization problem is proved NP-hard. This computational complexity is justified by two main factors. The first factor is about the important size of social networks. Modern social networks like TikTok and Facebook have reached an unprecedented number of users. Dynamic social networks, whose topology or/and informational content is able to evolve, represent the second factor. Maximizing influence in such networks remains a significant task. In this light, several methods have been proposed. Being motivated by this fact, we provide in this paper a detailed survey of influence maximization approaches. Our main concern is to provide a taxonomy of existing models in both static and dynamic networks. In addition, we provide a comparison of the state-of-the-art approaches according to a clear categorization. New trends for detecting influential nodes are also discussed. We provide then some challenges as well as future directions.

Social networks have attracted a great deal of attention and have, in fact, changed the way we produce, consume, and diffuse information. This change gave rise to the notion of social influence and today we talk about influential nodes. The process of detecting influential nodes in social networks aims to find entities that propagate information to a large portion of the network users. This process is often known as the influence maximization (IM) problem. Due to the explosive growth of social networks’ data, their structure is more complex and we talk about “big graph data.” Moreover, modern networks are dynamic and their topology or/and information is likely to change over time. Detecting influential nodes in such networks is a challenging task. Several methods have been developed in this context. However, they concentrate on static networks and there is little work on large-scale social networks. We propose in this article a new model for IM called MapReduce-based dynamic selection of influential nodes (MR-DSINs) that has the ability to cope with the huge size of real social networks. In fact, our approach is based on a graph sampling step in order to reduce the network’s size. Given that reduced version, MR-DSIN is able to select dynamically influential nodes. Our proposal has the advantage of considering the dynamics of information that can be modeled by users’ social actions (e.g., “share,” “comment,” “retweet”). Experimental results on real-world social networks and computer-generated artificial graphs demonstrate that MR-DSIN is efficient for identifying influential nodes, compared with three known proposals. We prove that our model is able to detect in the reduced graph an influence as important as in the original one.

In the medical field, a doctor must have a comprehensive knowledge by reading and writing narrative documents, and he is responsible for every decision he takes for patients. Unfortunately, it is very tiring to read all necessary information about drugs, diseases and patients due to the large amount of documents that are increasing every day. Consequently, so many medical errors can happen and even kill people. Likewise, there is such an important field that can handle this problem, which is the information extraction. There are several important tasks in this field to extract the important and desired information from unstructured text written in natural language. The main principal tasks are named entity recognition and relation extraction since they can structure the text by extracting the relevant information. However, in order to treat the narrative text we should use natural language processing techniques to extract useful information and features. In our paper, we introduce and discuss the several techniques and solutions used in these tasks. Furthermore, we outline the challenges in information extraction from medical documents. In our knowledge, this is the most comprehensive survey in the literature with an experimental analysis and a suggestion for some uncovered directions.

In this paper, we deal with the problem of community detection in attributed networks. The multi-objective grey wolf optimizer is adopted in order to discover the optimal partition with multiple objectives for the first time. This paper presents MoGoCo-@Net, a novel multiQbjective discrete grey wolf optimizatlon algorithm to solve the community detection-problem in the @ttributed networks. To fully exploit the topology structure and node attribute of vertices at each time step, we introduce two new criteria and maximize them simultaneously. Moreover, MoGoCo-@Net used opposition-based learning and improved label propagation technique by combining the node's attributes with the network topology for fast and effective initialization. Then, MoGoCo-@Net redefined the social hierarchy and the hunting behavior of grey wolves in discretization. Next, a multi-individual mutation operation is adopted as an evolutionary operation. In our experiments, various benchmark attributed networks are used to compare with some state-of-the-art methods. The experimental results show that the proposed algorithm performs favorably against the compared methods.

Social networks allow efficient interchange of ideas and informations which makes them important dissemination platforms. Due to the explosive use of such networks in our everyday life, their structure becomes more and more complexe. Consequently, modern social networks have reached important sizes. In fact, finding models that are able to scale over large data remains a daunting challenge. Reducing the social network's size is a key task in social netwoks analysis to deal with this data complexity. Many models have been adopted to deal with this problem. One of the most widely used methods for reducing the network's size is graph sampling, which corresponds to find a representative pattern from the initial graph while preserving its original properties. Choosing a subset of nodes or edges from the original graph is the simplest way to form a sample. In this paper, we aim to sample large scale social networks using a node sampling strategy. Our strategy is based on the distributed MapReduce paradigm to deal with huge networks. We evaluate our model using real social networks and we show that it can scale and it provides samples that preserve original networks' properties

Since the development of information technologies, there is a huge amount of electronic documents that was written by medical specialists and are rich of useful information needed to make critical decisions in several medical tasks. Thus, a doctor must have a big knowledge and he is responsible for every decision he takes for patients. In fact, the doctor should read, with full concentration, many electronic narrative documents to collect the necessary information. Unfortunately, it's too tiring to read all necessary information about drugs, diseases and patient due to the large amount of documents that are increasing every day. Consequently, so many medical errors can happen and even can cause fatalities. On the other hand, information extraction is such a good field that can handle this problem. One of the most important main task in this field is the Named Entity Recognition (NER) and its role was to identify the medical named entities, such as drug, disease or treatment, from an unstructured text written in natural language. However, in order to treat the narrative text, natural language tasks should be performed before NER. In our paper, we introduce a named entity recognition method, called NESSMa (Named Entity tagging by Surrounding Sequence Matching), and based on sequence tagging, it is able to annotate the words of a sentence using Bidirectional Long Short-Term Memory neural network with Conditional Random Field (BiLSTM-CRF) model. We pass the Bidirectional Encoder Representations from Transformers (BERT) word embedding as feature together with the Part of Speech (PoS) of the word and the cue sequence information. The cue sequence information indicates if a word belongs to a named entity surrounding sequence based on word edit distance. For that, we have automatically constructed a dictionary of named entities’ surrounding sequences for each entity type using a train set. As expected, experiments shows that adding the cue sequence information is able to improve the results according to F1-measure and outperform state-of-the-art methods.

Topic analysis (also called topic detection, topic modeling, or topic extraction) is a machine learning technique that organizes and understands large collections of text data, by assigning “tags” or categories according to each individual text's topic or theme. Topic analysis uses natural language processing (NLP) to break down human language into blocks (speech, words, sentences, context) so that you can find patterns and unlock semantic structures within texts to extract insights and help make data-driven decisions. The two most common approaches for topic analysis with machine learning are NLP topic modeling and NLP topic classification. Topic modeling faces several challenges: some are general to the NLP task (as extracting the context of document), and some are specific and are related to the nature (or properties) of the documents. One particular type of documents that raises several challenges are short-text documents that we find in Social Networks. First, we should note that with the growth of online social network platforms and applications, large amounts of textual user-generated content are created daily in the form of comments, reviews, and short-text messages. However, despite their ubiquity, extracting topics from shorts texts remains a difficult task for several reasons. First, unlike traditional normal texts, short texts typically only include a few words. Therefore, directly applying traditional models on short texts will suffer from the severe data sparsity problem (i.e., the sparse word co-occurrence patterns in individual document). Second, the limited contexts make it more difficult to identify the senses of ambiguous words in short texts. Third, in general, the performance of the used machine learning models relies on labeled data. Unfortunately, due to their volume, labeling short-text from social networks remains a tedious and hard task. In this paper, we propose a model for extracting topics in short-texts, and more specifically in Twitter. The key feature of our proposal is the use of word embedding technique for topic modeling, and k-Means clustering for the semi-automatic annotation of tweets. In addition, and unlike most existing approaches, we assign in our model a set of topics to a tweet each with a confidence degree.

Several approaches have been proposed to study opinions on Social Network Sites (SNS). Unfortunately, those works are not topic-sensitive and do not investigate the impact of emojis on text-based classification. In this paper, we propose a novel approach to predict the users’ opinions expressed through textual tweets and emojis. Thus, we construct an emoji sentiment lexicon. Then, we extract opinions from the text before considering both the text and emojis to see how they enhance the expression of opinions in SNS discussions. We conduct a set of benchmarks using several well-known machine learning algorithms, leading to an accuracy of 83, 7%.

Twitter is a communication medium and a collaboration system that allows broadcasting short messages called tweets. In contrast to traditional blogs, media-sharing and social networks services, microblogs (tweets) are textual messages submitted in real-time to report an idea, an actual interest, or an opinion. The size of these messages may be limited by a maximum number of characters. This constraint, related to the size of message, causes the use of a particular vocabulary. The aim is to exchange a maximum of information in as little characters as possible. In this respect, we will focus on the Tweet Contextualization task. The purpose of this task is to allow the reader a better understanding of the tweet. This paper deals with a new Tweet Contextualization approach using a semantic query expansion. The main idea of our proposed method is to enhance queries (tweets) in order to produce more informative contexts. The effectiveness of our method is proved through an experimental study conducted on the INEX 2014 collection.

Service-oriented computing (SOC) has come into view as a new computing paradigm that makes systems development and business process management more flexible and agile. It has recourse to services functioning as a fundamental compound in order to assist the construction of rapid and easy composition of distributed applications in heterogeneous environments. The rapid development of web service technologies has dramatically increased the number of available web services including those available on social networks. Thus, many studies have been carried out in the discovery mechanisms. The majority of proposed approaches in the literature for web services discovery based on the description of web services themselves and neglect their interaction with each other. Through the overlapping between social computing (illustrated by social networks) and service oriented computing (illustrated by web services), social web services are produced that are different from regular web services. In fact, they are a member of a social network and have an idea about other services in the same network. The first purpose of this paper is to justify the need of the use of an extension of WSDL-S to facilitate web service discovery, and the second purpose is to build a social network of web services to take advantage of the interactions that occur between services. However, the effort of developing social web services can partly undermine the utility of their implementation.

The great success of social networks is due to their ability to offer Internet users a space of free expression where they can produce a large amount of information which provides every day with a new challenge for data analysts. The ease of use of social media encourages users to increasingly express their opinions either by using simple words expressing feelings or by using irony and sarcasm. The new challenges are to extract and analyze this mass of information which can then be used in different applications such as sentiment analysis and sarcasm detection. Sarcasm detection is a subarea of sentiment analysis, opinion mining, and emotion mining which are all representing the process of automatic identification of people’s orientation or sentiment toward individuals, products, services, issues, and events. Sarcasm detection, which is the fact of deciding if a text is ironic or not, could be used, for example, to improve the precision of the sentiment analysis. In most of the existing approaches, sarcasm detection is a binary classification; each text is classified as sarcastic or non-sarcastic; however, since tweets are generally written by humans and humans are by default fuzzy in their emotions and expressions, we can’t 100% confirm that a text is sarcastic or not. In addition, tweets are expressed in natural language which is full of ambiguity and non-precision, which motivates us more to adopt fuzzy logic, not just to detect sarcasm but to give it a score. In this manuscript, we propose a fuzzy sarcasm detection approach using social information such as replies, historical tweets and likes, etc multiplying each by a degree of importance. The evaluation shows that the use of fuzzy logic has led us to improve the precision metric of the classification and to improve the accuracy of our approach. Using degrees of importance gave us the best values for recall, precision, and accuracy measures compared to existing approaches.

Detecting the influential nodes in dynamic social networks presents a very recent field that has gained considerable interest from researchers. One interesting approach is to update influential nodes incrementally taking into consideration the structural evolution of the social networks. However, most of the existing methods can only be used to identify influential nodes in static rather than dynamic social networks. In order to solve this problem, we propose an incremental approach for detecting influential nodes by inspecting social networks evolution. First, we identify the influential nodes in the original network. Then, we propose a method for finding the changed elements. Finally, we present our algorithm for updating influential nodes in dynamic social networks. Experimental results on three real dynamic social networks prove that our approach achieves better performance in terms of both influence degree and computational time.

With the increase of information on the web and online users’ activities, to find the appropriate information at the right time and discover items that customers are interested in among the available choices become difficult challenges. Recommender Systems have been introduced to overcome this problem by offering potentially relevant elements and providing users with personalized recommendations. Although several social-based recommendation techniques have been proposed in the literature, applying the community detection techniques based on previous transactions of users can improve the precision of the algorithms and better handle the data sparsity and cold-start problem. In this paper, we propose an improved association rule mining process based on Rule Power Factor to generate potent rules and enhance the accuracy of recommendation. The performance of the algorithm is analyzed on transactional datasets of MovieLens. The experimental results show that our method outperforms several state-of-the-art recommendation methods with increased precision, accuracy and minimum Mean Average Error values.

Retrieving relevant information from Twitter is always a challenging task given its vocabulary mismatch, sheer volume and noise. Representing the content of text tweets is a critical part of any microblog retrieval model. For this reason, deep neural networks can be used for learning good representations of text data and then conduct to a better matching. In this paper, we are interested in improving both representation and retrieval effectiveness in microblogs. For that, a Hybrid-Deep Neural-Network-based text representation model is proposed to extract effective features’ representations for clustering oriented microblog retrieval. HDNN combines recurrent neural network and feedforward neural network architectures. Specifically, using a bi-directional LSTM, we first generate a deep contextualized word representation which incorporates character n-grams form FasText. However, these contextual embedded existing in a high-dimensional space are not all important. Some of them are redundant, correlated and sometimes noisy making the learning models over-fitting, complex and less interpretable. To deal with these problems, we proposed a Hybrid-Regularized-Autoencoder-based method which combines autoencoder with Elastic Net regularization for an effective unsupervised feature selection and extraction. Our experimental results show that the performance of clustering and especially information retrieval in microblogs depend heavily on features’ representation.

Social networks have attracted a great deal of attention and have in fact important information vectors that have changed the way we produce, consume and diffuse information. Social networks' analysis has been of great interest and has encompassed different research areas including community detection, the discovery of web services from social networks, information diffusion, detection of infuential nodes. The process of detecting influential nodes in social networks is often khown as Influence Maximization (IM) problem, it deals with finding a small subset of nodes that spread maximum influence in the network. It has been proved that it has many applications such as the propagation of opinions, the study of the acceptance of political blogs or the study of the degree of adhesion of an actor to a product in marketing (web marketing). A such maximization requieres the presence of a diffusion model that controls information propagation within active individuals. This paper aims to provide a survey on the influence maximization problem and focuses on two aspects, influence diffusion models and proposed approaches for influential nodes detection. We start by describing formally the IM problem, then we will provide the state-of-the-art of both diffusion models and influence maximization algorithms.

Tweet contextualization (TC) is a new issue that aims to answer questions of the form What is this tweet about? The idea of this task was imagined as an extension of a previous area called multi-document summarization (MDS), which consists in generating a summary from many sources. In both TC and MDS, the summary should ideally contain most relevant information of the topic that is being discussed in the source texts (for MDS) and related to the query (for TC). Furthermore of being informative, a summary should be coherent, i.e. well written to be readable and grammatically compact. Hence, coherence is an essential characteristic in order to produce comprehensible texts. In this paper, we propose a new approach to improve readability and coherence for tweet contextualization based on bipartite graphs. The main idea of our proposed method is to reorder sentences in a given paragraph by combining most expressive words detection and HITS (Hyperlink- Induced Topic Search) algorithm to make up a coherent context.

Today, Twitter has become one of the most popular micro-blogging service with a large amount of information on various topics produced by millions of users every day. When searching for useful information in Twitter, users need to assess high quality content that meets their needs. However, the study of effective information retrieval in such microblog is still a challenge because there is a large difference in the quality level of relevant tweets returned in the search results for a given query. Therefore, looking for an effective microblog retrieval requires distinguishing the high-quality tweet content among thousands of results. Existing works are based on hand-crafted features (e.g., number of re-tweets, number of followers, etc.) using hard-hand-engineering to indicate the quality of tweets. In this paper, we focus on the problem of ranking tweets, and particularly on retrieving high quality content. We propose a ranking approach based on k-means clustering to distinguish high quality from low quality tweets. The clustering algorithm is based on learning features from deep learning autoencoder and hand-crafted features from tweets’ content and authors’ profiles. We used information gain as a feature importance measure to find the optimal set of features having stronger power in clustering the data. By conducting a pilot feature analysis study, we demonstrate the impact of the learned features to identify tweets’ quality in the clustering process. Our experimental results show that the integration of learned features has shown significant improvement in the quality of clustering and especially on the ranking performance compared to the use of hand-crafted features only.

Content-based 3D shape retrieval (CBR) become a central research topic in the field of pattern recognition. The primary challenge for CBR methods is to extract proper features for discriminating diverse shapes of 3D models with non-rigid deformations. The bag-of-features (BoF) paradigm and Local Feature Descriptors (LFDs) have demonstrated impressive levels of performance. The success of this approach prompted several extensions to BoF, such as sparse coding approach. In this study, state-of-the-art content-based 3D shape retrieval methods based on local features and sparse coding are investigated in detail. Performance of the well-known local spectral descriptors (GPS, HKS, SIHKS, WKS and meshSIFT) is tested with several algorithms for dictionaries learning and coefficient learning. Experiments are performed on the standard dataset Shrec11 with different scenarios.

The identification of central nodes has been a key problem in the field of social network analysis. In fact, it is a measure that accounts the popularity or the visibility of an actor within a network. In order to capture this concept, various measures, either sample or more elaborate, has been developed. Nevertheless, many of “traditional” measures are not designed to be applicable to huge data. This paper sets out a new node centrality index suitable for large social network. It uses the amount of the neighbors of a node and connections between them to characterize a “pivot” node in the graph. We presented experimental results on real data sets which show the efficiency of our proposal.

Community detection in social networks is a computationally challenging task that has attracted many researchers in the last decade. Most of approaches in the literature focus only on modeling structural properties, ignoring the social aspect in the relations between users. Additionally, they detect the communities, one after another, in a serial manner. However, the size of actual real-world social networks grows exponentially which makes such approaches inefficient. For this, several models tend to parallelize the community detection task. Unfortunately, social networks data often exhibits a high degree of dependency which renders the parallelization task more difficult. To overcome this difficulty, amongst the proposed distributed community detection methods, the label propagation algorithm (LPA) emerges as an effective detection method due to its time efficiency. Despite this advantage in computational time, the performance of LPA is affected by randomness in the algorithm. Indeed, LPA suffers from poor stability and occurrence of monster community. This paper introduces a new LPA algorithm for distributed community detection based on evidence theory which has shown a high efficiency in handling information. In our model, we will use the belief functions in the update of labels as well as in their propagation in order to improve the quality of the solutions computed by the standard LPA. The mass assignments and the plausibility, in our model, are computed based on the social influence for detecting the domain label of each node. Experimentation of our model on real-world and artificial LFR networks shows its efficiency compared to the state of the art algorithms.

Detecting influential nodes in social networks represents an essential issue for various applications to identify users that may maximize the influence of information in such networks. Several methods have been proposed to solve this problem often khown as influence maximization problem. However, most of them focused on the structure of the network and ignored the semantic aspect. Besides, these methods are parametric, they require the number k of influential elements in a deterministic manner. In this paper, we propose a parameterless algorithm called DIN (Detecting Influential Nodes in social networks) that combines the structure and the semantic aspect. The main idea of our proposal is to detect communities with overlap, modelize the semantic of each community then select influential elements. Experimental results on computer-generated artificial graphs demonstrate that DIN is efficient for identifying influential nodes, compared with two newly known proposals.

Over the last decades, there has been a growing interest to Service Oriented Computing (SOC) especially in a mobile environement. Web Services are the most common form of services for implementing SOC. Service Composition is to create a new business process by composing several services in order to fulfill business goals that individual services cannot achieve. The massive use of web services in the mobile devices imposes more manipulation of these services such as discovery, invocation, composition and execution. Web Service Composition in a mobile environement raises several challenges that does not exist in a classic (non-mobile) environement. Among these challenges is the change of context mainly due to the mobility of the smart device. We will refer to this in the paper as “uncertain context”. In this work we limit uncertain context to the couple(Location, bandwidth). We build an uncertain context based model, calculating the probability of context change. Then we try to find how the uncertain context influences service availability. This is traduced by the defined sensitivity function. Once showing how much does the services are sensitive to uncertain context, we build a web service composition model, based on web service sensitivity and presenting service composition plans according to their probabilities.

Malware detection has been one of the current computer security topics of great interest. Traditional signature-based malware detection fails to detect variants of known malware or previously unseen malware. To deal with this issue, machine learning and data mining methods have been widely used to counter the obfuscation techniques of attackers by examining the underlying behavior of suspected malware. However, these methods still suffer from the large number of extracted features and the lack of precise specifications which affects badly scanning time and the accuracy of the malware detection process. In this paper, we present an automatic detection method based on graph mining techniques. Maximal frequent subgraphs in a set of code graphs, representing common behaviors with precise specifications in execution files, are extracted and used as features to generate semantic signatures. These semantic signatures are represented by a set of learning models and employed to distinguish malware programs from benign. Experimental results indicate that our method extracts a limited number of interesting features and achieves effective malware detection.

Several approaches have been proposed for the problem of identifying authoritative actors in online communities. However, the majority of existing methods suffer from one or more of the following limitations: (1) There is a lack of an automatic mechanism to formally discriminate between authoritative and nonauthoritative users. In fact, a common approach to authoritative user identification is to provide a ranked list of users expecting authorities to come first. A major problem of such an approach is the question of where to stop reading the ranked list of users. How many users should be chosen as authoritative? (2) Supervised learning approaches for authoritative user identification suffer from their dependency on the training data. The problem here is that labeled samples are more difficult, expensive, and time consuming to obtain than unlabeled ones. (3) Several approaches rely on some user parameters to estimate an authority score. Detection accuracy of authoritative users can be seriously affected if incorrect values are used. In this article, we propose a parameterless mixture model-based approach that is capable of addressing the three aforementioned issues in a single framework. In our approach, we first represent each user with a feature vector composed of information related to its social behavior and activity in an online community. Next, we propose a statistical framework, based on the multivariate beta mixtures, in order to model the estimated set of feature vectors. The probability density function is therefore estimated and the beta component that corresponds to the most authoritative users is identified. The suitability of the proposed approach is illustrated on real data extracted from the Stack Exchange question-answering network and Twitter.

In this paper, we propose an agent-based approach for modeling dynamic connections in social networks. The key contribution of our work is the definition of a similarity measure for computing the strength of relationships between the social network members. For this purpose, we take into consideration the members' properties, the topological structure of the network and information about the interchange between each connected pair. To well choose the properties of social members, we use the concept of homophily. We show that our approach improves the effectiveness of the community detection process.

Content-Based Image Retrieval (CBIR) is becoming one of the most vivid research area in computer vision. It is widely used in medical applications especially in computer aided diagnostic systems (CAD). CBIR systems in digital mammography take an important part of these works. The work presented in this paper aims to propose a CBIR approach based on inexact graph matching algorithm for mammographic images. To achieve this task, we represent a mammogram as an Attributed Relational Graph (ARG) based on ImageMap approach where each node of the graph represents a semantic object. Objects that are considered in mammogram are: Background, Breast, Pectoral Muscle, Masses and Calcifications. Then, for each node, we compute a signature that describes the selected object. In order to retrieve the most similar images to a query one, a graph matching technique is applied based on the Hungarian algorithm. To Evaluate our approach 100 mammographic images from the MIAS database were used and six metrics were computed. Experiments demonstrate that the proposed method using Hamming distance gives the most promising results.

Abductive reasoning (or abduction) is the process of inferring hypotheses from observed data using a certain ‘knowledge’ encoded in the form of inference rules (or causal relations). Many important kinds of intellectual tasks, including medical diagnosis, fault diagnosis, scientific discovery, legal reasoning, and natural language understanding have been characterised as abduction. Unfortunately, abduction is 𝒩𝒫-hard. Genetic algorithms and biologically motivated computational paradigms inspired by the natural evolution turned out to be efficient in solving many hard problems while other existing approaches failed to solve in general. In this article, we present a genetic algorithm called HAKIM, for solving abduction problems. We encode an explanation in a chromosome-like structure, where every gene models a possible single hypothesis. Thereafter, we develop a fitness function that characterises the overall ‘quality’ of a chromosome representing an explanation; and then use standard genetic operators to compute a set of hypotheses that best explains the observed data. Simulation results on large-scale medical problems reveal the good performance of our model HAKIM.

Data mining (DM) is a new emerging discipline that aims to extract knowledge from data using several techniques. DM turned out to be useful in business where the data describing the customers and their transactions is in the order of terabytes. In this paper, we propose an approach for building customer models (said also profiles in the literature) from business data. Our approach is three-step. In the first step, we use fuzzy clustering to categorize customers, i.e., determine groups of customers. A key feature is that the number of groups (or clusters) is computed automatically from data using the partition entropy as a validity criteria. In the second step, we proceed to a dimensionality reduction which aims at keeping for each group of customers only the most informative attributes. For this, we define the information loss to quantify the information degree of an attribute. Hence, and as a result to this second step, we obtain groups of customers each described by a distinct set of attributes. In the third and final step, we use backpropagation neural networks to extract useful knowledge from these groups. Experimental results on real-world data sets reveal a good performance of our approach and should simulate future research.

In this work, we present a framework for the semantic composition of web services based on Statecharts and uniform community service descriptions. Our model is a two step process. In the first step, we derive the execution model of the user's query. The execution model is specified in Statecharts formalism; whereas the user's query is described in OWL-S. Therefore, a mapping from Statecharts formalism to OWL-S is developed. In the second step, we instantiate the developed execution model through invocation of available e-services instances. Hence, and a result, we obtain an execution plan (said also strategy) satisfying user constraints. The key features of the proposed framework could be summarized as follows. First, unlike other existing languages, using OWL-S enables the semantic description of e-services. These semantics are taken into consideration in our composition strategy. Second, the user constraints (or preferences) are taken into account during composition and are expressed as a finite set of logical formulas with the Knowledge Interchange Format (KIF) language.

This work extends a recent model for neural-based abductive reasoning to account for the monotonic class. A problem is said to be monotonic some causes, together, explain the same effect. For this, we developed a new computational principle, called the softmin, and implemented it within a neural architecture. Simulation results are very satisfactory and should stimulate future research.

Causal reasoning is a hard task that cognitive agents perform reliably and quickly. A particular class of causal reasoning that raises several difficulties is the cancellation class. Cancellation occurs when a set of causes (hypotheses) cancel each other's explanation with respect to a given effect (observation). For example, a cloudy sky may suggest a rainy weather; whereas a shiny sky may suggest the absence of rain. In this work we extend a recent neural model to handle cancellation interactions. Simulation results are very satisfactory and should encourage research.

Clustering is an important research area that has practical applications in many fields. Fuzzy clustering has shown advantages over crisp and probabilistic clustering, especially when there are significant overlaps between clusters. Most analytic fuzzy clustering approaches are derived from Bezdek's fuzzy c-means algorithm. One major factor that influences the determination of appropriate clusters in these approaches is an exponent parameter, called the fuzzifier. To our knowledge, no theoretical reason leading to an optimal setting of this parameter is available. This paper presents the development of an heuristic scheme for determining the fuzzifier. This scheme creates close interactions between the fuzzifier and the data set to be clustered. Experimental results in clustering IRIS data and in code book design required for image compression reveal a good performance of our proposal.