Paper: Feature Selection For Trainable Multilingual Broadcast News Segmentation

ACL ID N04-4023
Title Feature Selection For Trainable Multilingual Broadcast News Segmentation
Venue Human Language Technologies
Session Short Paper
Year 2004
Authors

Indexing and retrieving broadcast news stories within a large collection requires automatic de- tection of story boundaries. This video news story segmentation can use a wide range of au- dio, language, video, and image features. In this paper, we investigate the correlation be- tween automatically-derived multimodal fea- tures and story boundaries in seven different broadcast news sources in three languages. We identify several features that are important for all seven sources analyzed, and we discuss the contributions of other features that are impor- tant for a subset of the seven sources.