中文搜索引擎的搜索结果重合率研究

Study of Results Overlap and Uniqueness Among Major Chinese Web Search Engines

本文的研究目的是测试主流中文搜索引擎搜索结果之间的重合程度和差异程度。利用一个具有11 171条来自真实用户的提问样本集对百度、谷歌和中国雅虎进行实际测试,发现中文搜索引擎搜索结果之间的差异很大,重合率很低。在全部的第一页搜索结果中,三个引擎中任何一个引擎独有的搜索结果总数占89.34%,任何两个引擎之间重合的搜索结果总数占8.11%,三个引擎重合的搜索结果数量占2.54%。三个引擎前两页搜索结果的重合比例更低。通过和已有的英文搜索引擎重合率测试数据相比较,发现中英文搜索引擎的搜索结果重合率都很低,且很相近。

Based on a large number of queries, this study is to measure the overlap and differences among search results across three most popular Chinese web search engines, which are Baidu, Google. cn and Yahoo. cn. The set of randomly selected user-entered queries has 11 171 queries. Findings show that the percent of total first page results unique to only one of the three web search engines was 89.34 %, shared by two of the three web search engines 8.11%, and shared by all three web search engines 2.54 % . This small degree of overlap shows the significant difference in the way major Chinese web search engines retrieve and rank results as regards to the given queries. The research then compares these results with the overlap and uniqueness across the major English web search engines measured by American scholars before. Comparisons reflect that the overlap across major Chinese web search engines is the same as that across major English ones.