| 基于改进Single-Pass算法的BBS热点话题发现 |
| |
| 引用本文： | 马国栋,李慧.基于改进Single-Pass算法的BBS热点话题发现[J].首都师范大学学报(自然科学版),2014(6):13-17. |
| |
| 作者姓名： | 马国栋 李慧 |
| |
| 作者单位： | 首都师范大学教育技术系; |
| |
| 摘 要： | 详细介绍了经典增量式聚类算法Single-Pass的特点,针对该算法中"聚类质心不唯一"的缺陷提出"设定唯一聚类质心"的改进方案,降低了算法的时间复杂度;针对"算法聚类中心随机性强"的不足提出了"不断优化聚类中心"的改进方案,使得聚类中心的代表性更强.最后,基于"中华网BBS"文本数据对改进后的算法进行验证、分析,并与经典Single-Pass算法的聚类结果进行对比,得到了更为理想的实验结果,表明算法可行有效.
|
| 关 键 词： | Single-Pass 文本聚类 热点话题发现 |
Hot Topic Detecting in BBS with Developed Single-Pass |
| |
| Ma Guodong Li Hui.Hot Topic Detecting in BBS with Developed Single-Pass[J].Journal of Capital Normal University(Natural Science Edition),2014(6):13-17. |
| |
| Authors: | Ma Guodong Li Hui |
| |
| Institution: | Ma Guodong Li Hui (Department of Educational Technology, Capital Normal University, Beijing 100048) |
| |
| Abstract: | Introducing the feature of Single-Pass which is a classical clustering algorithmof that ＂ set a unique cluster centroid＂. And I complexity of the centers＂. Finally improved the program for the algorithm of ＂cluster centroid is not the only＂, reducing the time algorithm. I also improved the algorithm defect of ＂ algorithmic randomness strong cluster I got a data results to prove the improved algorithm is reliable. |
| |
| Keywords: | single-pass text clustering hot topic detection |
| 本文献已被 CNKI 维普 等数据库收录！ |
