Sites Inria

Version française

ORPAILLEUR Research team

Knowledge representation, reasonning

Team presentation

The Orpailleur Team is mainly interested in Knowledge Discovery in Databases (KDD) and in Knowledge Engineering (KE). KDD consists in processing large volumes of data for discovering patterns that are significant and reusable. Considering patterns as gold nuggets and databases as locations to be explored, KDD can be likened to the process of searching for gold. This explains the name of the research team, as, in French, ``orpailleur'' denotes a gold miner.

KDD is an iterative and interactive process based on three main operations: data preparation, data mining and interpretation of the extracted patterns. Domain knowledge can be used for improving and guiding the KDD process, leading to Knowledge Discovery guided by Domain Knowledge or KDDK. The discovered patterns can be represented as knowledge units using a knowledge representation formalism and integrated within a knowledge base for problem-solving needs. Knowledge discovery and knowledge engineering are two complementary processes supporting the research lines in the Orpailleur Team

Application domains investigated by the Orpailleur team are related to life sciences, such as agronomy, biology, chemistry and medicine. Cooking, cultural heritage, network quality ans security, web of data (Linked Data) are other application domains of interest.

Research themes

The research topics of the ORPAILLEUR project-team are the following:

  • Symbolic methods in knowledge discovery: pattern mining, sequence mining, graph mining, text mining, biclustering, functional dependencies, redescription mining.
  • Formal Concept Analysis and extensions: pattern structure, relational concept analysis, triadic analysis.
  • Numerical methods in knowledge discovery: hidden Markov models, deep learning, graphical networks, hybrid data mining, meta-mining.
  • Knowledge Engineering: semantic web technologies, web of data, ontology engineering, formalization of reasoning, revision.
  • Decision making, aggregation theory, preferences, rough sets, multi-valued logic.

International and industrial relations

  • European Project CrossCult.
  • ANR Projects: Elker, PractikPharma.
  • Mastodons QCM-Biochem (Data Quality and Complexity).
  • International collaborations: Brazil, Canada, Chili, Czech Republic, Russia, Spain, Uruguay, USA.

Keywords: Knowledge Discovery Data Mining Knowledge Engineering