A rule-based chinese question answering system for reading comprehension tests

Abstract

We have developed a rule-based system, which is called Chinese question answering system for reading comprehension (Cqarc). This paper systematically introduces and discusses key techniques in the Cqarc, describes the processing mechanism of the Cqarc, presents processing strategies and methods in the Cqarc. Cqarc uses heuristic rules to look for lexical and semantic clues in the question and the story. Finally we have tested Cqarc on Chinese reading comprehension corpus (CRCC). Overall, Cqarc found the correct sentence 42% of the time, which is encouraging given the simplicity of its rules. To the best of our knowledge, this is the first work that reports that the use of a rule-based approach achieves satisfying results on answering questions for Chinese reading comprehension tests.