聚类搜索引擎探究  

Study on Clustering Search Engine

作　　者：黄建年[1,2] 侯汉清[1] 

机构地区：[1]南京农业大学 [2]南京财经大学

出　　处：《图书馆学研究》2009年第1期32-35,共4页Research on Library Science

摘　　要：一批代表性的聚类搜索引擎的出现,使得聚类搜索引擎的优劣逐步为公众所知。采用聚类算法、重视结果的显示方式、关注用户提问信息、提供个性化服务均是聚类搜索引擎最显著的优点,应该引起足够的重视。针对过分依托原生搜索引擎、没有形成专用聚类算法、搜索速度相对较慢、聚类层次有限等现状,作者最后还对聚类搜索引擎的发展提出了一些建议,比如实现人工聚类与自动聚类结合、增加学术趋势分析功能、加强相关度研究等。Emergence of clustering search engine makes their advantages and disadvantages known by public. We should pay special attentions to these advantages such as adopting clustering algorithm, visualization of search results, taking notice of user query, supporting personalized service and providing relative tools. As far as some disadvantages, such as completely relying on other general search engines, limited clustering algorithm, slower search speed and limited clustering levels, are concerned, the author brings forward some recommendations about development of clustering search engine, such as integrating artificial clustering with computer clustering, supporting analysis of academic trend, and strengthening a study of relativity.

关 键 词：搜索引擎 聚类搜索 超级搜索 自动聚类 动态聚类 

分 类 号：G354[文化科学—情报学]