JAVELIN III: Answering Non-Factoid Questions in Japanese

In this paper, we describe our adaptation of the JAVELIN system to Japanese question answering for the NTCIR-6 QAC track. To establish a baseline Japaneseto-Japanese non-factoid question answering system, we performed the minimum extensions to our factoid question answering system. The answer boundary recognition task was simplified by introducing a “One Sentence Assumption” so that answer extraction task only deals with ranking of answer candidates in one sentence level. In the end, the performance of our machine learning-based sub-modules was affected by a scarcity of training data; nevertheless, our system performed close to the average accuracy in the number of questions correctly answered.