Abstract | ||
---|---|---|
Our work addresses automatic detection of enunciations and segments with reformulations in French spoken corpora. The proposed approach is syntagmatic. It is based on reformulation markers and specificities of spoken language. The reference data are built manually and have gone through consensus. Automatic methods, based on rules and CRF machine learning, are proposed in order to detect the enunciations and segments that contain reformulations. With the CRF models, different features are exploited within a window of various sizes. Detection of enunciations with reformulations shows up to 0.66 precision. The tests performed for the detection of reformulated segments indicate that the task remains difficult. The best average performance values reach up to 0.65 F-measure, 0.75 precision, and 0.63 recall. We have several perspectives to this work for improving the detection of reformulated segments and for studying the data from other points of view. |
Year | Venue | Keywords |
---|---|---|
2016 | LREC 2016 - TENTH INTERNATIONAL CONFERENCE ON LANGUAGE RESOURCES AND EVALUATION | Spoken Corpora,Reformulation,Reformulation Marker,Paraphrase,Supervised Machine Learning |
Field | DocType | Citations |
Computer science,Syntagmatic analysis,Speech recognition,Paraphrase,Natural language processing,Artificial intelligence,Recall,Spoken language | Conference | 0 |
PageRank | References | Authors |
0.34 | 16 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
N. Grabar | 1 | 32 | 7.31 |
Iris Eshkol-Taravella | 2 | 4 | 5.23 |