Abstract:
With the abundance of information available today, we need ecient tools to explore it. Search engines try to retrieve the most relevant documents for a given query, but still require users to look for the exact answer. Question Answering (Q&A) systems go one step further by trying to answer users' questions posed in natural language. In this paper we describe the semantic approach to Q&A retrieval applied for Bulgarian language. We are investigating how the usage of named entity recognition, question answer type detection and dependency parsing can improve the retrieval of answer-bearing structures compared to the bag-of-words model. Moreover, we evaluate nine dierent dependency parsing algorithms for Bulgarian, and a named entity recognizer trained with data automatically extracted from Wikipedia.