%0 Conference Paper %1 zhao08characerizing %A Zhao, Qiankun %A Bhowmick, Sourav S. %A Zheng, Xin %A Yi, Kai %B CIKM '08: Proceeding of the 17th ACM conference on Information and knowledge management %C New York, NY, USA %D 2008 %I ACM %K behavior community detection dynamic evolution mining modeling %P 309--318 %R http://doi.acm.org/10.1145/1458082.1458125 %T Characterizing and predicting community members from evolutionary and heterogeneous networks %U http://portal.acm.org/citation.cfm?id=1458082.1458125&coll=GUIDE&dl=GUIDE&CFID=17695029&CFTOKEN=22803053 %X Mining different types of communities from web data have attracted a lot of research efforts in recent years. However, none of the existing community mining techniques has taken into account both the dynamic as well as heterogeneous nature of web data. In this paper, we propose to characterize and predict community members from the evolution of heterogeneous web data. We first propose a general framework for analyzing the evolution of heterogeneous networks. Then, the academic network, which is extracted from 1 million computer science papers, is used as an example to illustrate the framework. Finally, two example applications of the academic network are presented. Experimental results with a real and very large heterogeneous academic network show that our proposed framework can produce good results in terms of community member recommendation. Also, novel knowledge and insights can be gained by analyzing the community evolution pattern. %@ 978-1-59593-991-3

@inproceedings{zhao08characerizing, abstract = {Mining different types of communities from web data have attracted a lot of research efforts in recent years. However, none of the existing community mining techniques has taken into account both the dynamic as well as heterogeneous nature of web data. In this paper, we propose to characterize and predict community members from the evolution of heterogeneous web data. We first propose a general framework for analyzing the evolution of heterogeneous networks. Then, the academic network, which is extracted from 1 million computer science papers, is used as an example to illustrate the framework. Finally, two example applications of the academic network are presented. Experimental results with a real and very large heterogeneous academic network show that our proposed framework can produce good results in terms of community member recommendation. Also, novel knowledge and insights can be gained by analyzing the community evolution pattern.}, added-at = {2009-09-21T13:52:25.000+0200}, address = {New York, NY, USA}, author = {Zhao, Qiankun and Bhowmick, Sourav S. and Zheng, Xin and Yi, Kai}, biburl = {https://www.bibsonomy.org/bibtex/282f379a50b22c0fc84ed306c19576c7c/taynaud}, booktitle = {CIKM '08: Proceeding of the 17th ACM conference on Information and knowledge management}, description = {Characterizing and predicting community members from evolutionary and heterogeneous networks}, doi = {http://doi.acm.org/10.1145/1458082.1458125}, interhash = {d363574c5a5df4b8af0d3e143d23d033}, intrahash = {82f379a50b22c0fc84ed306c19576c7c}, isbn = {978-1-59593-991-3}, keywords = {behavior community detection dynamic evolution mining modeling}, localfile = {/home/aynaud/biblio/articles/CIKM-08-Network.pdf}, location = {Napa Valley, California, USA}, pages = {309--318}, publisher = {ACM}, timestamp = {2009-09-21T13:52:29.000+0200}, title = {Characterizing and predicting community members from evolutionary and heterogeneous networks}, url = {http://portal.acm.org/citation.cfm?id=1458082.1458125\&coll=GUIDE\&dl=GUIDE\&CFID=17695029\&CFTOKEN=22803053}, year = 2008 }