基于聚类分析的Web信息搜索算法研究  

The Web Information Search Algorithm Research Based on Cluster Analysis

作　　者：刘顺来[1] 

机构地区：[1]广州航海高等专科学校计算机与信息工程系,广东广州510725

出　　处：《电脑与电信》2007年第6期53-56,共4页Computer & Telecommunication

摘　　要：本文分析了Web信息搜索中的Leader算法其聚类结果中可能存在重叠区域;k-means算法其k参数的确定可能需要多次尝试。结合上述二种聚类分析算法提出将Leader算法和k-means算法融合可以同时解决上述问题,且在实际的Web信息搜索中更能体现聚类分析的优势。Analyzes Web in the information search in the Leader algorithm its duster result possibly to have the overlap region; k-means algorithm its k parameter determination possibly needs to attempt many times. Unifies above two kind of duster parsing algorithm to propose may simultaneously solve the Leader algorithm and the k-means algorithm fusion the above question, also can manifest the duster analysis in the actual Web information search the superiority.

关 键 词：WEB信息搜索 Leader算法 K-MEANS算法 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]