论文部分内容阅读
【目的】提高问答系统准确率,提升目前问答系统的满意度。【应用背景】在自然语言处理领域,问答系统已成为一个重要研究热点,但现阶段问答系统的准确率较低。【方法】对智能聊天机器人ALICE源码进行分析研究,并对其进行二次开发,加入中文分词,在其内部推理分析的基础上,提出一种利用本体上下位关系对用户查询问题的推荐方法。【结果】将领域本体库融入到ALICE机器人中,对用户问题进行分析,并从中提取关键词,从本体库中查询所提取的相关关键词,得出与用户提问相匹配的答案并推荐给用户。【结论】实验结果证明,加入本体的推荐结果后,用户满意度有较大提升。
【Objective】 To improve the accuracy of question answering system and enhance the satisfaction of current question answering system. [Background] In the field of natural language processing, question answering system has become an important research hotspot, but the accuracy of question answering system is low at this stage. 【Method】 The ALICE source code of intelligent chatting robot is analyzed and researched, and its secondary development is carried out. The Chinese participle is added. Based on its internal reasoning analysis, a recommendation method is proposed based on on-and-off bit on user query. [Results] The domain ontology library was integrated into ALICE robot, the user problems were analyzed, and the key words were extracted from the ontology base, the extracted relevant keywords were obtained and the answers matching the questions of the users were obtained and recommended to the users . 【Conclusion】 The experimental results show that user satisfaction is greatly improved after adding ontology recommendation results.