LDA主题发现java源码

LDAGibbsSampling ================ /** Copyright (C) 2013 by SMU Text Mining Group/Singapore Management University/Peking University LDAGibbsSampling is distributed for research purpose, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. Feel free to contact the following people if you find any problems in the package. lyang@cs.umass.edu * */ Brief Introduction =================== 1. This is Liu Yang's implementation for Gibbs Sampling of LDA. The test data set is Newsgroup-18828, which is included in the project. You can test other data sets with it. Just import the project into Eclipse and run LdaGibbsSampling.java to start it without any configuration. The original documents and sample output files have been included. 2. Author's technical blog : http://blog.csdn.net/yangliuy Author's homepage:https://people.cs.umass.edu/~lyang For more information of LDA and Gibbs Sampling: http://blog.csdn.net/yangliuy/article/details/8302599 3. This is a initial implementation for the Topic Expertise Model which is proposed in the following paper: Liu Yang, Minghui Qiu, Swapna Gottipati, Feida Zhu, Jing Jiang, Huiping Sun and Zhong Chen. CQARank: Jointly Model Topics and Expertise in Community Question Answering. In Proceedings of the 22nd ACM International Conference on Information and Knowledge Management (CIKM 2013). (http://dl.acm.org/citation.cfm?id=2505720) If you use this model implementation, please cite this paper. 4. We will also release more open source code for topic models in https://github.com/yangliuy.