Query based summarization using topic background knowledge

A query based summary is intended to retrieve a summary of information from a document of a given query. It is useful to help users understand the main information biases of queries. Typically, queries lack semantic information, so the query based summary effect is not ideal. In this paper, we propose a new method based on theme background knowledge to solve problems. In particular, by adding several search engines search query subject related documents to build theme background knowledge, and then the PageRank algorithm is applied to the specified document contains background knowledge and information. The experimental results demonstrate the effectiveness of the proposed method.