The widespread availability of electronic documents and the exponential growth of the World Wide Web have made the automatic categorization of documents a critical method for organizing information and facilitating knowledge discovery, and information retrieval. This paper aims to examine the key techniques and methodologies utilized in multilingual document classification, while also bringing attention to some of the complex challenges that still need to be addressed. In particular, the paper presents a thorough review of the literature concerning the theory and methods of multilingual document representation and classification.

Topic Modeling (TM) can act as a bridge linking unstructured text data to a structured knowledge representation in a Knowledge Graph (KG). The Latent Dirichlet Allocation (LDA) is a commonly used distributional term clustering technique in this regard. However, existing distributional term clustering approaches have not utilized LDA as a bottom-up training strategy with prior knowledge to cluster semantically related terms as concepts across different domains for building and enhancing a GSCKG. We propose to employ a Twice-Trained Agglomerative hierarchical framework using LDA over Generic Semantic Core KG (T2AggLDA-GSCKG), outlined in five steps. We aim to align term topics with the predefined Core Concepts (CCs) of SCKGs, thereby designating modules of a GSCKG that facilitate both its building and enrichment. Our goal is to adapt the LDA term clustering process by utilizing a topic seed-based LDA model to consider these CCs. During the 1stLDA training, we endeavor to identify hypernym and related relations among noun phrase patterns to construct SCKG, and during the 2ndtraining, we aim to enhance it. To achieve this, we will boost the 2ndtraining LDA input data by benefiting from CCs’ two prior knowledge techniques for topic-seed terms incorporation namely seed key knowledge injection and entity masking. The evaluation results show that our proposal has an overwhelming term clustering performance over CCs. It outperforms other unsupervised and semi-supervised distributional baselines on two datasets related to fish hunting and ontology domains, with nearly 13 times higher precision compared to that of normal LDA training.

Due to a large number of documents available on the web, operations such as finding a set of information contained in a document has become a difficult task, especially with multilingual documents. Hence the necessity to have performance tools for finding, organizing and classifying information. A variety of classification methods are proposed to resolve this kind of problem but these techniques suffer from limits such as the loss of information, and the loss of relations between words that affects the effectiveness and the performance of the classification process. So, this paper attempts to support the idea of multilingual document classification, especially in the biomedical domain using a new approach, based on deep learning. The key idea is to generate a new conceptual representation of textual multilingual medical documents to facilitate the classification task. In this context, a deep learning technique will be exploited for a good representation. To show the feasibility of our approach, we implemented a system related to a domain that attracts more and more attention from the data mining community: the biomedical domain. An experimental study is performed, using documents extracted from the biomedical benchmark corpus, called Oshumed, which contains documents distributed by different categories.

Recently, the social network has revolutionized the interaction and information exchange between users. Several works have been committed to unify the social network domain, particularly the medical domain through introducing ontology-based modeling of medical social network. Nevertheless, few researches focused on modeling the social network using ontology from the medical side. To overcome this drawback, we propose, in this paper, a medical social network ontology which includes definitions of main entities and describes major attributes of medical social network concepts aiming at sharing common understanding of this domain.

We will focus in this paper on the competitive intelligence problem which deals with the competitive environment of a company. Our purpose is to predict and anticipate the action of its competitor. We are talking here about a context of reasoning under uncertainty. All existed works define the concept of competitive intelligence and propose a scheme for the competitive intelligence process and its stages, but there is no work, at the best of our knowledge, that touched the practical aspect of the field or developed a complete competitive intelligence solution that can be delivered to the decision maker, which makes the originality of our work. To motivate the research, we will address a competitive practical case in the field of telecommunications. In this paper we propose a competitive intelligence solution composed by two steps: actions association using k-modes algorithm which has the capability to deal with nominal data, and actions generation using rough set theory which has the capability to deal with inexact data and drive rules from it.

In this paper, we aim to cover works that are related to the process of transforming requirements into UML diagrams, from the first works which were manual techniques in 1976, to automatic tools in 2015. In this context, we try to exhibit different approaches and to indicate their strength as well as their shortcomings. This work will help us to evaluate existing approaches and propose other alternatives for Requirement Engineering. The objective of this paper is to present an overview of various works dedicated to requirement analysis and a comparative study of these works. Also, we tried to discuss the combination of Artificial Intelligence with Requirement Engineering.

Summary Software development life cycle is a structured process, including the definition of user requirements specification, the system design, and programming. The design task comprises the transfer of natural language specifications into models. The class diagram of Unified Modeling Language has been considered as one of the most useful diagrams. It is a formal description of user's requirements and serves as inputs to the developers. The automated extraction of UML class diagram from natural language requirements is a highly challenging task. This paper explains our vision of an automated tool for class diagram generation from user requirements expressed in natural language. Our new approach amalgamates the statistical and pattern recognition properties of natural language processing techniques. More than 1000 patterns are defined for the extraction of the class diagram concepts. Once these concepts are captured, an XML Metadata Interchange file is generated and imported with a Computer‐Aided Software Engineering tool to build the corresponding UML class diagram. Copyright © 2015 John Wiley & Sons, Ltd.

Competitive intelligence deals with the competitive environment of a company. Several studies have been conducted on competitive intelligence domain but there is no empirical work that gives a complete implemented competitive intelligence solution. This paper presents an overview of competitive intelligence studies and highlights the issues towards developing a complete CI solution. A new conceptual model which details the collection phase and incorporates the anticipation of the competitor decisions is proposed. To better solve the CI problem, the paper proposes to integrate the ability of multi-criteria decision aid methods to manage conflicting criteria in a complex environment, with the ability of artificial intelligence in managing and extracting large amount of technical data/information in such a context.

During the past decade, the advent of the social network has offered several platforms that promote communication among users on common spaces. Several efforts were devoted to unify the social network domain, particularly the scientific domain through introducing ontology-based modeling of scientific social network. However, the measurement of the researchers standings within the scientific community is generally absent. To overcome this drawback, we propose, in this paper, a scientific social network ontology which includes definitions of main entities and describes main attributes of : Scientific social network concepts aiming to share common understanding of this domain and to reflect the academic career paths.

Decision making has become a strategic need for any business. Indeed, it is among the priorities of capital business. The establishment of decision information systems facilitates the data exploitation and analysis. We distinguish data warehouses as the core system of business intelligence to ensure the structuring and analysis of multidimensional data. Consequently, the design of data warehouses has become a major problem, leading to the development of appropriate approaches to implement data warehouses. In this paper, we propose an approach to design and to construct data warehouses based on a descriptive statistics technique for the analysis of multidimensional data in the Principal Components Analysis (PCA). The findings of this article appear in two main areas: (i) a conceptual model data warehouse, (ii) an algorithm for the determination of measures and dimensions. A case study is used to validate our proposal.

Recently, Information and Communication Technologies have introduced new practices in human resource management functions such as e-recruitment. Job seekers submit their Curriculum Vitae (CV) via the Web, or send them directly to a company. The area of e-recruitment is facing a growing number of these documents which are in different formats, and contain a large amount of information. Then, it has become imperative to use automated techniques to identify, extract, and exploit information from CVs to find the most appropriate one for a given post. Our work focuses on CVs analysis. We present a system for analyzing and structuring CVs which are in French language. For his end, we made an extension of General Architecture of Text Engineering (GATE) by formulating necessary rules that generate new annotations. The goal is to normalize the CV content according to the structure adopted by Europass CV. This action is guided by the HR-XML standard. An empirical study is conducted to validate the proposed process and we show that there is an improvement in the extraction phase.