一种基于带权集合的搜索引擎隐式反馈算法-论文.pdfVIP

Computer Science and Application 计算机科学与应用, 2011, 1, 128-133 doi:10.4236/csa.2011.13026 Published Online December 2011 (/journal/csa) A Weighted Set-Based Implicit Feedback Algorithm for Search Engine * Hui Zhang , Yan Chen State Key Lab. of Software Development Environment, Beihang University, Beijing Email: {hzhang, chy}@ Received: Oct. 7th, 2011; revised: Oct. 29th, 2011; accepted: Nov. 19th, 2011. Abstract: With the quick development of the Internet, online information recourses are becoming richer and richer. However, traditional search engines are becoming hard to acquire the users’ retrieval intention because of the short keywords user inputted, so it is difficult to satisfy the needs of the user. In this paper, we proposed a Weighted Set-based implicit feedback algorithm to improve accuracy of the search engines effectively. Th- rough analyzing the characteristics of the snippets of Web pages search engine returned, we present a kind of Weighted-Set for representing Web page snippets and a method for calculating the weight of elements in this Set. In addition, we design an approach to calculate the intersection between any two Weighted-Sets, which can obtain users’ implicit search intention automatically, so that we can improve search accuracy in the way of query expansion. Query expansion experiments on the popular Google search engine show that our algo- rithm can improve search accuracy effectively. Keywords: Weighted-Set; Search Engine; Implicit Feedback; Intersection Operation 一种基于带权集合的搜索引擎隐式反馈算法 张 辉* ，陈 岩 北京航空航天大学软件开发环境国家重点实验室，北京 Email: {hzha