Multi-Label Text Classification with a Mixture Model Trained by EM

@misc{citeulike:2049150, abstract = {In many important document classication tasks, documents may each be associated with multiple class labels. This paper describes a Bayesian classication approach in which the multiple classes that comprise a document are represented by a mixture model. While the labeled training data indicates which classes were responsible for generating a document, it does not indicate which class was responsible for generating each word. Thus we use EM to ll in this missing value, learning both the...}, author = {Mccallum, Andrew K. }, citeulike-article-id = {2049150}, keywords = {classification, em, model, text}, posted-at = {2008-01-28 05:31:03}, priority = {2}, title = {Multi-Label Text Classification with a Mixture Model Trained by EM}, url = {http://citeseer.ist.psu.edu/202622.html} }

TY - GEN ID - citeulike:2049150 L3 - citeulike-article-id:2049150 TI - Multi-Label Text Classification with a Mixture Model Trained by EM N2 - In many important document classication tasks, documents may each be associated with multiple class labels. This paper describes a Bayesian classication approach in which the multiple classes that comprise a document are represented by a mixture model. While the labeled training data indicates which classes were responsible for generating a document, it does not indicate which class was responsible for generating each word. Thus we use EM to ll in this missing value, learning both the... KW - classification KW - em KW - model KW - text AU - Mccallum, Andrew UR - http://citeseer.ist.psu.edu/202622.html ER -