Olivier Teste

Analyse multidimensionnelle de documentsviades dimensions OLAP

Avec l’emergence de formats de donnees semi-structures (tels que XML), le stockage de documents dans un entrepot centralise est apparu de facon naturelle comme une adaptation des entrepots de donnees. De nos jours, les systemes OLAP (On-Line Analytical Processing) font face a une part grandissante de donnees non numeriques. Cet article presente un environnement pour l’analyse multidimensionnelle de donnees textuelles dans un environnement OLAP. La structure, les metadonnees et le contenu des documents orientes texte sont transposes en sujets d’analyse (faits) et en axes d’analyse (dimensions) au sein d’un schema en etoile modifie. Ceci permet de plus amples possibilites d’analyses multidimensionnelles. Cet environnement permet a un utilisateur d’avoir une vision detaillee au sein d’une collection de documents.

Querying Heterogeneous Document Stores

NoSQL document stores offer support to store documents described using various structures. Hence, the user has to formulate queries using the possible representations of the desired information from different schemas. In this paper, we propose a novel approach that enables querying operators over a collection of documents with structural heterogeneity. Our work introduces an automatic query rewriting mechanism based on combinations of elementary operators: project, restrict and aggregate. We generate a custom dictionary that tracks all representations for attributes used in the documents. Finally, we discuss the results of our approach with a series of experiments.

BigDimETL with NoSQL Database

Abstract In the last decade, we have witnessed an explosion of data volume available on the Web. This is due to the rapid technological advances with the availability of smart devices and social networks such as Twitter, Facebook, Instagram, etc. Hence, the concept of Big Data was created to face this constant increase. In this context, many domains should take in consideration this growth of data, especially, the Business Intelligence (BI) domain. Where, it is full of important knowledge that is crucial for effective decision making. However, new problems and challenges have appeared for the Decision Support System that must be addressed. Accordingly, the purpose of this paper is to adapt Extract-Transform-Load (ETL) processes with Big Data technologies, in order to support decision-making and knowledge discovery. In this paper, we propose a new approach called Big Dimensional ETL (BigDimETL) dealing with ETL development process and taking into account the Multidimensional structure. In addition, in order to accelerate data handling we used the MapReduce paradigm and Hbase as a distributed storage mechanism that provides data warehousing capabilities. Experimental results show that our ETL operation adaptation can perform well especially with Join operation.

POMap: An Effective Pairwise Ontology Matching System.

The identification of alignments between heterogeneous ontologies is one of the main research issues in the semantic web. The manual matching of the ontologies is a complex, time consuming and an error prone task. Therefore, ontology matching systems aims to automate this process. Usually, these systems perform the matching process by combining element and structural level matchers. Selecting the optimal string similarity measure associated with its threshold is an important issue in order to enhance the effectiveness of the element level matcher, which in turn will improve the whole ontology system results. In this paper, we present POMap, an ontology matching system based on a syntactic study covering element and structural levels. For the element level matcher we have adopted the best configuration based on the analysis of the performances of many string similarity measures associated with their thresholds. For the structural level, we have performed a syntactic study on both subclasses and siblings in order to infer the structural similarity. Our proposed matching system is validated and evaluated on the Anatomy, the Conference and the Large Biomedical tracks provided by the benchmark of OAEI 2016 ontology matching campaign.