Readitopics: Make Your Topic Models Readable via Labeling and Browsing

Julien Velcin (1) , Antoine Gourru (1) , Erwan Giry-Fouquet (1) , Christophe Gravier (2) , Mathieu Roche (3) , Pascal Poncelet (4)

1 ERIC - Entrepôts, Représentation et Ingénierie des Connaissances
2 LHC - Laboratoire Hubert Curien
3 UMR TETIS - Territoires, Environnement, Télédétection et Information Spatiale
4 ADVANSE - ADVanced Analytics for data SciencE

Readitopics provides a new tool for browsing a textual corpus that showcases several recent work for labeling topic models and estimating topic coherence. We will demonstrate the potential of these techniques to get a deeper understanding of the topics that structure different kinds of datasets. This tool is provided as a Web demo but it can be easily installed to experiment with your own dataset. It can be further extended to deal with more advanced topic modeling techniques.

Domaines: Base de données [cs.DB]

Julien Velcin, Antoine Gourru, Erwan Giry-Fouquet, Christophe Gravier, Mathieu Roche, et al.. Readitopics: Make Your Topic Models Readable via Labeling and Browsing. IJCAI: International Joint Conference on Artificial Intelligence, Jul 2018, Stockholm, Sweden. ⟨lirmm-01910611⟩