| 一种基于流行度和中心度的内容网络社区发现方法 |
| |
| 引用本文： | 康钊宁,贾彩燕,柴变芳. 一种基于流行度和中心度的内容网络社区发现方法[J]. 中国科技论文在线, 2013, 0(10): 986-988,997 |
| |
| 作者姓名： | 康钊宁 贾彩燕 柴变芳 |
| |
| 作者单位： | [1]北京交通大学计算机与信息技术学院,北京100044 [2]交通数据分析与挖掘北京市重点实验室,北京100044 |
| |
| 基金项目： | 国家自然科学基金资助项目（60905029）;中央高校基本科研业务费专项资金资助项目（2011JBM025）;北京市自然科学基金资助项目（4112046）;河北省自然科学基金资助项目（F2013205192）;河北省教育厅科技计划资助项目（ZD2010128）致谢：非常感谢Prof.C.Lee Giles提供的Citeseer论文引证网络数据. |
| |
| 摘 要： | 基于近年来发展的社区发现概率模型的可解释性，对现有的性能较好的基于节点中心度和流行度的社区划分链接模型PPL进行扩展，给出了一种新的可以结合节点内容的组合模型PPL-IX2。该模型不但可解决节点属性的选择问题，并可充分利用节点间的链接关系。实验结果表明，新给出的PPL-DC模型优于单纯的链接模型及已有的链接和内容相结合的组合模型。
|
| 关 键 词： | 社区发现 链接模型 内容网络 |
A novel method for community detection on content networks based on node popularity and node productivity |
| |
| Kang Zhaoning,Jia Caiyan,Chai Bianfang. A novel method for community detection on content networks based on node popularity and node productivity[J]. Sciencepaper Online, 2013, 0(10): 986-988,997 |
| |
| Authors: | Kang Zhaoning Jia Caiyan Chai Bianfang |
| |
| Affiliation: | 1. School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China 2. Beijing Key Lab of Traffic Data Analysis and Mining ,Beijing 100044, China) |
| |
| Abstract: | Based on the interpretability of probability models on community detection, this paper extends the state-of-the-art prob- ability model PPL, which is based on node popularity and productivity, and presents a novel model PPL-DC which combines con- tent of nodes and links. PPL-DC model addresses the problem of attributes selection on nodes. It makes full use of links between nodes. The experimental results have shown that the proposed PPL-DC model outperforms the original link models and classical combination models which combine links and content of nodes. |
| |
| Keywords: | community detection link model content networks |
| 本文献已被 维普 等数据库收录！ |
