A Web-based Question Answering System for Effective e-Learning

Web based question answering (QA) system is a valuable tool for improving e-learning. Many approaches use natural language processing technology to understand questions, which is incomplete and error-prone. Besides, instead of extracting detailed answer, many approaches simply return hyperlinks to potential answers, which is inconvenient for the users. In this paper we use template mapping technique to detect the type of a question, based on which an appropriate query for a specific search engine is built. We then extract detailed content blocks from relevant pages as answers. The content blocks are detected using a projection technique and clustered based on similarity. We select the answer cluster using the redundancy property of Web answers. The content blocks in the answer cluster are put into one page for easy comparison. In this way we can greatly improve the effectiveness of Web QA systems for e-learning.