| 一种大规模中文搜索日志的层次聚类方法 |
| |
| 引用本文： | 孙锐,金澎.一种大规模中文搜索日志的层次聚类方法[J].科技通报,2012,28(8):83-85. |
| |
| 作者姓名： | 孙锐 金澎 |
| |
| 作者单位： | 乐山师范学院智能信息处理及应用重点实验室,四川乐山,614004 |
| |
| 基金项目： | 四川省教育厅青年基金资助 |
| |
| 摘 要： | 提出一种层次聚类算法,旨在对搜索引擎的查询日志数据进行聚类分析。算法基于搜狗实验室公开的查询日志数据,通过3次划分完成查询文本聚类,每一次划分实现不同程度的降维。相似度参数可根据不同的聚类需求调整,算法可扩展性强。实验结果为查询推荐、相关性排序等提供了有力的依据。
|
| 关 键 词： | 搜索引擎 查询日志 层次聚类 聚类分析 |
Hierarchical Clustering Method for Large-scale Chinese Query Logs |
| |
| SUN Rui
,
JIN Peng.Hierarchical Clustering Method for Large-scale Chinese Query Logs[J].Bulletin of Science and Technology,2012,28(8):83-85. |
| |
| Authors: | SUN Rui JIN Peng |
| |
| Institution: | (Laboratory of Intelligent Information Processing and Application Leshan Normal University,Leshan 614004,China) |
| |
| Abstract: | This paper propose a hierarchical clustering algorithm for search engine query log data.The algorithm is divided into three steps to accomplish clustering based on Sogou query log.The data dimensionality is variously reduced in each step.Similarity parameters can be adjusted according to the different requirement in the scalable algorithm.The experiment results provide a strong basis for query recommendation and relevance ranking. |
| |
| Keywords: | search engine query log hierarchical clustering cluster analysis |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
