Development of an advanced open-domain question answering system by incorporating text mining methods

点击翻译按钮获取中文摘要

Open domain question answering is a promising technology for obtaining the information relevant to one's information needs from large-scale documents. In this research, we focused on "mining type questions", which had not been dealt with the existing QA system. We firstly investigated the mining type questions appeared in the real-world question database, then developed a QA system targeting them by exploiting the findings. In addition, we developed the methods to improve the accuracy of the traditional question answering, the QA methods targeting a foreign language documents, and the methods targeting speech data.