Abstract: Existing methods for query precision of meta-search engine are not quite efficient.In order to provide better query precision for users，a new method is proposed which based on combining of topic information and correlation on the model of topic meta-search.This algorithm considers correlation between topic information and member search engine，correlation between topic information and search results，correlation between query word and search results，and repeatability of search results，position of results and other factors.Experiments show that the algorithm is a good way to put the results in front which are most concerned，and can improve the precision.

Key words: meta-search, topic, correlation, result conformity

摘要： 当前存在的元搜索查询方法不够精确，为了解决这一问题，建立了主题元搜索模型，提出了一种基于主题信息与相关度相结合的结果整合算法。该算法综合考虑了主题信息与成员搜索引擎之间的相关度，主题信息与查询结果之间的相关度，查询词与查询结果的相关度，以及查询结果的重复度、位置等诸多因素。实验表明，该算法能够很好地把用户最关心的结果排在查询结果集的前面，提高了查准率。

关键词: 元搜索, 主题, 相关度, 结果整合