Doctoral Consortium – Paper 17

Cleaning Noisy Knowledge Graphs Ankur Padia Doctoral Consortium October 22, 2017, 11:00. TC.3.07 Download paper (preprint)

Abstract

My dissertation research is developing an approach to identifying and correcting errors in a knowledge graph produced by extracting entities and relations from text. Information extraction systems can automatically construct knowledge graphs from a large collection of documents, which might be drawn from news articles, Web pages, social media posts or discussion forums. The language processing task is challenging and current extraction systems introduce many kinds of errors. Previous work on improving the quality of knowledge graphs using additional evidence mined from background knowledge ...