摘要
对传统搜索引擎使用的排序算法进行分析,将概率意义引入对象级搜索,提出一种在对象级搜索中合并多个查询结果的算法.通过归一化以及考虑概率意义与查询规模间的关系,该算法独立于生成结果列表使用的排序算法.实验结果证明该算法提高了查询结果的质量.
Through analyzing the ranking algorithms adopted by traditional search engines and introducing the idea of probability into object level search, a multi-object query merging algorithm, which can be used in object-level search engines, was proposed. The algorithm is independent of the ranking algorithm which generates the ranking lists by the way of performing normalization and considering the relationship between the probability sense and the number of queries. The experimental results show that the proposed algorithm improves the quality of query results.
基金
国家自然科学基金(60573077)
高等学校博士学科点专项科研基金(2007105)
教育部新世纪优秀人才支持计划(NCET-05-0549)资助
关键词
搜索引擎
对象级搜索
排序算法
多查询合并
search engine
object-level search
ranking algorithm
multi-query merging