一种基于词共现图的受限领域自动问答系统

提出了一种基于高频词和共现词的文本主题词抽取方法。该方法充分考虑到文档的统计信息和语义信息, 通过对提问问题和答案库中答案的相似度计算排序, 输出候选答案。提出一种具体的应用模型, 分别从问题的分析、信息检索和答案抽取三个模块进行系统的设计, 具有一定的应用价值。

System of automatic question-answering based on term co-occurrence

This paper provided a text keyphrase extraction method based on term co-occurrence and high frequency words. The method utilized the statistical information and semantic information of the document, output the candidate answer sorted by the similarity of the question and the answer in the answer library. It presented a specific application model which had a certain value, designed analysis of the problem, information retrieval and answer extraction modules.