Cascading Behavior in Large Blog Graphs Export

@article{citeulike:2940613, abstract = {How do blogs cite and influence each other? How do such links evolve? Doesthe popularity of old blog posts drop exponentially with time? These are someof the questions that we address in this work. Our goal is to build a modelthat generates realistic cascades, so that it can help us with link predictionand outlier detection.Blogs (weblogs) have become an important medium of information because oftheir timely publication, ease of use, and wide availability. In fact, theyoften make headlines, by discussing and discovering evidence about politicalevents and facts. Often blogs link to one another, creating a publiclyavailable record of how information and influence spreads through an underlyingsocial network. Aggregating links from several blog posts creates a directedgraph which we analyze to discover the patterns of information propagation inblogspace, and thereby understand the underlying social network. Not only areblogs interesting on their own merit, but our analysis also sheds light on howrumors, viruses, and ideas propagate over social and computer networks.Here we report some surprising findings of the blog linking and informationpropagation structure, after we analyzed one of the largest available datasets,with 45,000 blogs and \~{} 2.2 million blog-postings. Our analysis also shedslight on how rumors, viruses, and ideas propagate over social and computernetworks. We also present a simple model that mimics the spread of informationon the blogosphere, and produces information cascades very similar to thosefound in real life.}, archivePrefix = {arXiv}, author = {Leskovec, Jure and Mcglohon, Mary and Faloutsos, Christos and Glance, Natalie and Hurst, Matthew}, citeulike-article-id = {2940613}, citeulike-linkout-0 = {http://arxiv.org/abs/0704.2803}, citeulike-linkout-1 = {http://arxiv.org/pdf/0704.2803}, day = {20}, eprint = {0704.2803}, keywords = {behavior, blogging, blogosphere, online, social-network, web20}, month = {Apr}, posted-at = {2009-10-09 07:17:19}, priority = {2}, title = {Cascading Behavior in Large Blog Graphs}, url = {http://arxiv.org/abs/0704.2803}, year = {2007} }

TY - JOUR ID - citeulike:2940613 L3 - citeulike-article-id:2940613 N2 - How do blogs cite and influence each other? How do such links evolve? Doesthe popularity of old blog posts drop exponentially with time? These are someof the questions that we address in this work. Our goal is to build a modelthat generates realistic cascades, so that it can help us with link predictionand outlier detection.Blogs (weblogs) have become an important medium of information because oftheir timely publication, ease of use, and wide availability. In fact, theyoften make headlines, by discussing and discovering evidence about politicalevents and facts. Often blogs link to one another, creating a publiclyavailable record of how information and influence spreads through an underlyingsocial network. Aggregating links from several blog posts creates a directedgraph which we analyze to discover the patterns of information propagation inblogspace, and thereby understand the underlying social network. Not only areblogs interesting on their own merit, but our analysis also sheds light on howrumors, viruses, and ideas propagate over social and computer networks.Here we report some surprising findings of the blog linking and informationpropagation structure, after we analyzed one of the largest available datasets,with 45,000 blogs and ~ 2.2 million blog-postings. Our analysis also shedslight on how rumors, viruses, and ideas propagate over social and computernetworks. We also present a simple model that mimics the spread of informationon the blogosphere, and produces information cascades very similar to thosefound in real life. TI - Cascading Behavior in Large Blog Graphs KW - behavior KW - blogging KW - blogosphere KW - online KW - social-network KW - web20 AU - Leskovec, Jure AU - Mcglohon, Mary AU - Faloutsos, Christos AU - Glance, Natalie AU - Hurst, Matthew PY - 2007/Apr/20/ UR - http://arxiv.org/abs/0704.2803 ER -