Algorithmes de classification et d'optimisation: participation du LIA/ADOC á DEFT'14
Luis Adrián Cabrera-Diego, Stéphane Huet, Bassam Jabaian, Alejandro Molina, Juan-Manuel Torres-Moreno, Marc El-Bèze, Barthélémy Durette
arXiv:1702.06510v1
Originality Synthesis-oriented
AI Analysis
This work addresses a specific text mining challenge for the DEFT campaign, but it is incremental as it applies existing methods to a new dataset.
The paper tackled the task of identifying the session in which articles from previous TALN conferences were presented, using three statistical systems, and achieved a micro-precision score of 0.76 on the test corpus.
This year, the DEFT campaign (Défi Fouilles de Textes) incorporates a task which aims at identifying the session in which articles of previous TALN conferences were presented. We describe the three statistical systems developed at LIA/ADOC for this task. A fusion of these systems enables us to obtain interesting results (micro-precision score of 0.76 measured on the test corpus)