Self-Supervised Relation Learning from the Web

Web extraction systems attempt to use the immense amount of unlabeled text in the Web in order to create large lists of entities and relations. Unlike traditional IE methods, the Web extraction systems do not label every mention of the target entity or relation, instead focusing on extracting as many different instances as possible while keeping the precision of the resulting list reasonably high. SRES is a Web relation extraction system that performs its task in a totally unsupervised way. It requires neither hand-written rules nor labeled training examples, but learns the extraction patterns from unlabeled text, using short descriptions of the target relations and their attributes. We further enhance...