Identifying Sentence-Level Semantic Content Units with Topic Models

Abstract--Statistical approaches to document content modeling typically focus either on broad topics or on discourselevel subtopics of a text. We present an analysis of the performance of probabilistic topic models on the task of learning sentence-level topics that are similar to facts. The identification of sentential content with the same meaning is an important task in multi-document summarization and the evaluation of multi-document summaries. In our approach, each sentence is represented as a distribution over topics, and each topic is a distribution over words. We compare the topicsentence assignments discovered by a topic model to goldstandard assignments that were manually annotated on a set of closely related pairs of news articles. We observe a clear correspondence between automatically identified and annotated topics. The high accuracy of automatically discovered topicsentence assignments suggests that topic models can be utilized to identify (sub-)sentential semantic conten...

• Type: Conference
• Year: 2010
• Where: DEXAW
• Authors: Leonhard Hennig, Thomas Strecker, Sascha Narr, Ernesto William De Luca, Sahin Albayrak
• Keywords: Database, DEXAW 2010, Probabilistic Topic Models, Topic Models, Topicsentence Assignments