Text Classification from Positive and Unlabeled Documents

@misc{citeulike:1877823, abstract = {Most existing studies of text classification assume that the training data are completely labeled. In reality, however, many information retrieval problems can be more accurately described as learning a binary classifier from a set of incompletely labeled examples, where we typically have a small number of labeled positive examples and a very large number of unlabeled examples. In this paper, we study such a problem of performing Text Classification WithOut labeled Negative data (TC-WON). In...}, author = {Yu, Hwanjo and Zhai, Chengxiang and Han, Jiawei }, citeulike-article-id = {1877823}, keywords = {categorization, classification, machines, positive, support, text, unlabelled, vector}, posted-at = {2007-11-07 15:12:38}, priority = {2}, title = {Text Classification from Positive and Unlabeled Documents}, url = {http://citeseer.ist.psu.edu/yu03text.html} }

TY - GEN ID - citeulike:1877823 L3 - citeulike-article-id:1877823 N2 - Most existing studies of text classification assume that the training data are completely labeled. In reality, however, many information retrieval problems can be more accurately described as learning a binary classifier from a set of incompletely labeled examples, where we typically have a small number of labeled positive examples and a very large number of unlabeled examples. In this paper, we study such a problem of performing Text Classification WithOut labeled Negative data (TC-WON). In... TI - Text Classification from Positive and Unlabeled Documents KW - categorization KW - classification KW - machines KW - positive KW - support KW - text KW - unlabelled KW - vector AU - Yu, Hwanjo AU - Zhai, Chengxiang AU - Han, Jiawei UR - http://citeseer.ist.psu.edu/yu03text.html ER -