摘要
相关查询可以给用户推荐合适的查询以辅助用户快速获得需要的信息，其中根据原始查询选择相关查询的优劣是相关查询有效性的关键。提出利用用户查询日志获得查询语义相关性，其关键是定义了三个与原始查询相关的相似性测度：查询串流行程度、查询串字面相似性和查询串之间语义相似性。使用真值程度度量，将查询串的三个测度度量值映射成一个查询相关性客观评价值，在计算过程中还能得到每个特性的优劣程度。实验结果表明：语义相关查询及其客观评价方法可以显著提高相关查询质量，主、客观评价值之间具有相当高的相关性，表明了本方法挖掘相关查询和评价相关查询的有效性。
Related queries can recommend the appropriate queries to help users quickly access the information needed. It is the key for the effectiveness of related queries to select the related queries. We bring forward a method to make use of the users＇query logs to obtain the semantic related queries, whose key is the definition of the three methods of similarity measure associated with the original query： Query prevalence, query literal similarity and semantic similarity between queries. We use the measurement of degree of truth scale to map the three evaluation values of query into an objective measurement of query relevance and we can also get the merits and demerits of each characteristic. The experimental results show that： the semantic-related queries and objective evaluation method can significantly improve the quality of related queries, and the values between subjective evaluation and objective evaluation are highly related, which means the effectiveness of our method mining and evaluating the related queries.
出处
《情报学报》
CSSCI
北大核心
2012年第6期574-582,共9页
Journal of the China Society for Scientific and Technical Information
基金
本研究得到桂林电子科技大学广西可信软件重点实验室开放基金、苏州大学江苏省计算机信息处理技术重点实验室基金（KJS0714）、武汉大学软件工程国家重点实验室开放基金（BJ211002）资助.
