Semantic Pattern based Dependency Matching for Exact Answer Retrieval

摘要

In this paper, we retrieve exact answer from a question/answer pair based on semantic pattern and dependency matching. Question target and the weight information of main phrase are gained from semantic pattern. Candidate exact answers to a question are retrieved from its corresponding sentence-form answer by dependency relation matching. A probabilistic model, in which weight information is included, is then adopted to select the most appropriate one. The semantic correctness of selected exact answer can be validated by question target. Our experiments achieve almost an accuracy of 70% in the open domain and 80% in the medicine domain.