Skip to Main content Skip to Navigation
Conference papers

Evaluating Automatic Topic Segmentation as a Segment Retrieval Task

Abstract : Several evaluation metrics have been proposed for topic seg-mentation. Most of them rely on the paradigm that seg-mentation is mainly a task that detects boundaries, and thus are oriented on boundary detection evaluation. Nevertheless, this paradigm is not appropriate to get homogeneous chapters, which is one of the major applications of topic segmentation. For instance on Broadcast News, topic segmentation enables users to watch a chapter independently of the others. We propose to consider segmentation as a task that detects homogeneous segments, and we propose evaluation metrics oriented on segment retrieval. The proposed metrics are experimented on various TV shows from different channels. Results are analysed and discussed, highlighting their relevance.
Document type :
Conference papers
Complete list of metadata

Cited literature [14 references]  Display  Hide  Download
Contributor : Yannick Estève Connect in order to contact the contributor
Submitted on : Tuesday, September 12, 2017 - 1:08:11 AM
Last modification on : Thursday, December 21, 2017 - 1:04:21 AM
Long-term archiving on: : Wednesday, December 13, 2017 - 1:34:50 PM


Files produced by the author(s)




Abdessalam Bouchekif, Delphine Charlet, Géraldine Damnati, Nathalie Camelin, yannick Estève. Evaluating Automatic Topic Segmentation as a Segment Retrieval Task. Interspeech 2017, Aug 2017, Stockholm, Sweden. pp.2924 - 2928, ⟨10.21437/Interspeech.2017-1231⟩. ⟨hal-01585769⟩



Record views


Files downloads