Hybrid Feature Factored System for Scoring Extracted Passage Relevance in Regulatory Filings - Naver Labs Europe
preloder

We report in this paper our contribution to the FEIII 2017 challenge addressing relevance ranking for information of passages extracted from 10-K and 10-Q regulatory fillings underlined by based on specific relationships they mention between mentioned financial entities and their role with respect to the reporting companies. We leveraged our previous works done on document structure and content analysis for regulatory fillings to train hybrid text analytics and decision making models. We designed and trained several layers of classifiers feed fed with linguistic and semantic features to improve relevance prediction. We present in this paper our experiments and the achieved results on the competition data set.

NAVER LABS Europe
NAVER LABS Europe
Ceci correspond à une petite biographie d'environ 200 caractéres