Statistical entity-topic models Export

@inproceedings{citeulike:821915, abstract = {ABSTRACT The primary purpose of news articles is to convey information about who, what, when and where. But learning and summarizing these relationships for collections of thousands to millions of articles is difficult. While statistical topic models have been highly successful at topically summarizing huge collections of text documents, they do not explicitly address the textual interactions between who/where, i.e. named entities (persons, organizations, locations) and what, i.e. the topics. We present new graphical models that directly learn the relationship between topics discussed in news articles and entities mentioned in each article. We show how these entity-topic models, through a better understanding of the entity-topic relationships, are better at making predictions about entities.}, address = {New York, NY, USA}, author = {Newman, David and Chemudugunta, Chaitanya and Smyth, Padhraic}, booktitle = {KDD '06: Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining}, citeulike-article-id = {821915}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1150402.1150487}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/1150402.1150487}, citeulike-linkout-2 = {http://www.ics.uci.edu/~newman/Newman_SIGKDD06.pdf}, comment = {Infer how entities relate to topics. Distinguish between normal words and entities (composed expressions). In this sense this is somehow related to the bigram topic model of Mc Callum and (? Teh?) but using a well established named entity recognizer instead of pure statistics. Nice toolbox of concepts/tricks in graphical model Application domain: news Exploitation for: Ontology mining }, doi = {10.1145/1150402.1150487}, isbn = {1595933395}, keywords = {entitytopicmodels, ontology, topicinference, topicmodels}, pages = {680--686}, posted-at = {2006-08-30 11:43:11}, priority = {0}, publisher = {ACM Press}, title = {Statistical entity-topic models}, url = {http://dx.doi.org/10.1145/1150402.1150487}, year = {2006} }

TY - CONF ID - citeulike:821915 L3 - citeulike-article-id:821915 N2 - ABSTRACT The primary purpose of news articles is to convey information about who, what, when and where. But learning and summarizing these relationships for collections of thousands to millions of articles is difficult. While statistical topic models have been highly successful at topically summarizing huge collections of text documents, they do not explicitly address the textual interactions between who/where, i.e. named entities (persons, organizations, locations) and what, i.e. the topics. We present new graphical models that directly learn the relationship between topics discussed in news articles and entities mentioned in each article. We show how these entity-topic models, through a better understanding of the entity-topic relationships, are better at making predictions about entities. SN - 1595933395 AD - New York, NY, USA EP - 686 TI - Statistical entity-topic models PB - ACM Press T2 - KDD '06: Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining SP - 680 KW - entitytopicmodels KW - ontology KW - topicinference KW - topicmodels N1 - Infer how entities relate to topics. Distinguish between normal words and entities (composed expressions). In this sense this is somehow related to the bigram topic model of Mc Callum and (? Teh?) but using a well established named entity recognizer instead of pure statistics. Nice toolbox of concepts/tricks in graphical model Application domain: news Exploitation for: Ontology mining AU - Newman, David AU - Chemudugunta, Chaitanya AU - Smyth, Padhraic PY - 2006/// UR - http://dx.doi.org/10.1145/1150402.1150487 ER -