基于用户兴趣模型的元搜索引擎算法研究

Algorithm research of meta search engine based on user interest model

为了解决当前人们在使用搜索引擎获取信息时遇到的搜索结果针对性不强,无关信息太多的问题,提出了将用户兴趣建模技术和元搜索引擎算法相结合的搜索引擎工作方式。首先收集用户的个人信息和常用浏览用于用户兴趣模型的构建,然后创建一种映射算法,能够把用户查询较好地映射到已创建的用户兴趣模型,用来计算与各成员引擎的相关度。在元搜索引擎的成员引擎特征表示上提出了以兴趣分类采样为依据,同时对成员引擎调度算法也进行改进,选择与当前用户查询相关度最高的几个成员引擎完成搜索工作。经过分析表明这种算法可以有效改进查询速度和精度。

currently when people use search engine to obtain information, there is too much irrelevant information in the search results. In order to solve the problem, we proposed a new search engine working method that combins user interest modeling techniques with meta search engine algorithm. First, we constructed user interest model based on User's information and browsing behavior. Second, we put forward a mapping algorithm that map the user query to user interest class, used to calculate the correlation with member engines. Then, we constructed member engine feature representation of meta search engine based on interest classification sampling, and improved member engine scheduling algorithm, so that meta search engine can select several member engines associated with the current user query to complete search. Analysis shows that this algorithm can effectively improve the search speed and precision.