SVM answer selection for open-domain question answering

Abstract

This paper presents an answer selection method based on Support Vector Machines (SVM) for Open-Domain Question Answering (QA). Selecting and ranking plausible answers from a large number of candidates in documents is one of the most critical parts of QA systems. It is extremely difficult to find good evaluation functions or rules for the answer selection. To overcome this issue, we apply SVM to answer selection. We evaluate the performance measured by mean reciprocal rank (MRR) and the correct ratio of answer ranked first. The results show that the proposed SVM-based method offers a statistically significant increase in performance compared to other machine learning methods such as decision tree learning (C4.5) boosting with decision tree learning (C5.0), and the maximum entropy method.