【摘要】 现有自动解题系统从试题文本及相关资源获取解题知识，但由于试题本身潜在的复杂性与多样性，所获取的知识可能不足以支持实际解题，需要进一步获取其他解题知识.本文基于三支概念分析理论，通过构造试题文本三支概念格和决策规则提取，进一步挖掘试题文本中隐含的解题知识，在所收集数据集上进行的试验结果表明，本方法有效提升了自动解题系统的解题性能.更多还原

【Abstract】 The existing automatic problem-solving system obtains relevant knowledge from problem texts and related resources. However, due to prospective complexity and diversity of the exam itself, there are a lot of limitations to the current question-answering method which requires further other knowledge in order to answer the exam questions. Based on the theory of three-way concept analysis, this paper constructs three-way concept lattices of problem texts and extracts decision rules to further mine implicit problem-solving knowledge. The experimental results on the collected data sets show that the proposed method can improve the performance of the automatic problem solving system.更多还原