基于HNC句类的社区问答系统问句检索模型构建

社区问答系统中充斥着大量的噪声，给用户检索信息造成麻烦，以往的问句检索模型大多集中在词语层面。针对以上问题构建句子层面的问句检索模型。新模型基于概念层次网络（hierarchincal network of concept，HNC）理论当中的句类知识，从句子的语用、语法和语义三个层面计算问句间相似度。通过问句分类算法确定查询问句和候选问句的问句类别，得到问句间的语用相似度，利用句类表达式的结构和语义块组成分别计算问句间的语法及语义相似度。在真实数据集上的实验表明，基于HNC句类的新模型提高了问句检索结果的准确性。

Construction of question retrieval model in community question answering system based on HNC sentence-category

Community question answering system causes trouble for users to retrieve information due to useless information. Most of the previous question retrieval models focus on the word level. In order to solve the above problems, this paper proposed a question retrieval model at the sentence level. Based on the sentence-category of HNC theory, the new model calculated similarities between questions from the pragmatic, grammatical and semantic levels of the sentence. The model used the question classification algorithm to determine the categories of query question and candidate question, and thus obtained pragmatic similarity between questions. It used the sentence expression structure and the sentence semantic block to calculate grammatical and semantic similarities. Experiments on real data sets show that the new model based on HNC sentence-category improves the accuracy of question retrieval results.