Detection and Analysis of Paraphrastic Reformulations in Spoken Corpora (Repérage et analyse de la reformulation paraphrastique dans les corpus oraux) [in French]

Our work addresses the automatic detection of paraphrastic rephrasing in spoken corpus. The proposed approach is syntagmatic. It is based on paraphrastic rephrasing markers and the specificities of the spoken language. Manual annotation performed by two annotators provides fine-grained and multi-dimensional description of the reference data. Automatic method is proposed in order to decide whether sentences contain or not the paraphrases. The obtained results show up to 66.4% precision. The analysis of the manual annotations indicates that there are few cases in which paraphrastic segments show morphological modifications (inflection, derivation or compounding) or syntactic equivalence. Mots-clés : Paraphrase, reformulation, corpus oral, marqueurs de reformulation paraphrastique.