Contextual Web Summarization: A Supervised Ranking Approach

Abstract

We investigate the task of reading context-biased web summarization, where the goal is to extract information relevant to the current reading context from a cited web article. In certain kind of linked document sets such as Wikipedia articles, scientific papers as well as news and blogs, such contextual summaries can be useful in providing additional related information to the user helping in the reading task. In this work, we focus on web articles only and try to find out the set of key components that contribute to building up the reading context. We build a supervised model for ranking sentences from the cited document according to their contextual salience. Initial evaluation based on annotated data-set of web articles show that our ranking model performs better than the generic summaries as well as baseline context-biased summaries.