基于向量空间模型的中文搜索引擎评测系统研究与实现*

提出了能够综合衡量搜索引擎工作性能的六个评测指标,选择了当今主流的三个搜索引擎：谷歌、雅虎和百度进行评测,成功地在两个大型的数据集上实现了整个自动评测系统。实验表明,谷歌工作性能最稳定;雅虎返回的第一个结果最能满足用户的要求,但会受时间因素的影响;百度明显地受关键字类别因素的影响。最后对中搜、狗、爱问进行了评测。

Research and implementation of Chinese search engines evaluation system based on VSM

This paper proposed six targets to evaluate the engines in a comprehensive manner. It focused on the evaluation on the performance of the most popular engines, such as Google, Yahoo and Baidu etc. The system was accomplished and implemented successfully on the two kinds of large scale experimental datasets prepared in advance. Experimental results indicate that Google performs most stably as a whole, Yahoo always returns the best first result while it is a little sensitive to the time factor, Baidu is obviously sensitive to the categories factor. Finally, the system was applied to the other three famous Chinese search engines listed as China search, Sogou and I ask to prove the applicability of this system.