Nos partenaires



Accueil du site > English > Research Topics > Topic 2 - Indexing and Information Search

Topic 2 - Indexing and Information Search

In this topic, the research issues we tackle concern :

  • querying and effectively analyzing raw factual and textual data, as well as pre-processed data,
  • displaying the most relevant information according to users’ requirements by optimizing the performances both in terms of effectiveness and efficiency.

To address these research issues, our research activities consist in defining methods and tools for modeling structured, semi-structured or unstructured data. We have identified 3 complementary targets :

  • First, our research aims to understand the variety of data collections, their distribution, their heterogeneity, and their large volumes of data and documents they represent.
  • Second, our research allows modeling users by representing their interests, their usage (profiles) and their environment (geographical, social, medical, ubiquitous ...).
  • Finally, our research provides solutions for modeling processes performed by these users on the information in different environments : centralized or large-scale distributed ones.

As a result, our research consists in a wide range of activities :

  • Expressing users’ requirements for information and accessing data and documents : adaptation to the context of queries, query disambiguation, query-based diversified graphics metaphors algebra for policy analysis ;
  • Contextual Information Retrieval : adaptive indexing and search models, evaluation protocols, selective information search... ;
  • Optimizing accesses : optimizing dynamic queries in large-scale distributed environments ;
  • Analyzing, mining and visualizing information : aggregating and integrating information, analyzing actor and information networks, applying business intelligence and scientometrics, modeling OLAP multi-aggregation.

Topic 2 is composed of 29 permanent positions (12 professors and 17 associate professors) and hosts each year ten new students preparing a Ph.D. thesis. These researchers belong to two teams : Pyramid and SIG.

Cette rubrique ne contient aucun article.