Summarizing Search Results with Community-Based Question Answering

Previous work on snippet generation focused mainly on how to produce one snippet for an individual search result. This paper aims to generate snippets as a comprehensive overview for an entity query (e.g., flu) in a search-result page. Our approach first extracts the attributes (e.g., Symptom and diagnose) of the categories (e.g., Disease) from a community-based question-answering (CQA) website, and then generates the snippets based on how central a sentence is to the meaning of the query, its category, and how well it diversifies the attributes. Integer Linear Programming (ILP) is adopted to find the optimal sentence set. The experiments are conducted on Wikipedia and Yahoo! Answers. Experimental results demonstrate the effectiveness of our approach, compared to an existing commercial search engine and several summarization baselines.