Document stream clustering : experimenting an incremental algorithm and AR-based tools for highlighting dynamic trends

Alain Lelu (1) , Martine Cadot (2) , Pascal Cuxac (3)

1 LASELDI - Laboratoire de Semio-Linguistique, Didactique et Informatique
2 PAROLE - Analysis, perception and recognition of speech
3 INIST - Institut de l'information scientifique et technique

We address here two major challenges presented by dynamic data mining: 1) the stability challenge: we have implemented a rigorous incremental density-based clustering algorithm, independent from any initial conditions and ordering of the data-vectors stream, 2) the cognitive challenge: we have implemented a stringent selection process of association rules between clusters at time t-1 and time t for directly generating the main conclusions about the dynamics of a data-stream. We illustrate these points with an application to a two years and 2600 documents scientific information database.

Mots clés: Fouille de texte; Data-stream clustering; Classification incrémentale; Règles d'association

Domaines: Intelligence artificielle [cs.AI]; Traitement du texte et du document

Alain Lelu, Martine Cadot, Pascal Cuxac. Document stream clustering : experimenting an incremental algorithm and AR-based tools for highlighting dynamic trends. International Workshop on Webometrics, Informetrics and Scientometrics & Seventh COLLNET Meeting, LORIA, May 2006, Nancy, France. pp.345-352. ⟨hal-00336175⟩