Entity Disambiguation with Web Links

Entity Disambiguation with Web Links Andrew Chisholm School of Information Technologies University of Sydney NSW 2006, Australia andy.chisholm.89@gmail.com Ben Hachey School of Information Technologies University of Sydney NSW 2006, Australia ben.hachey@gmail.com Abstract Entity disambiguation with Wikipedia relies on structured information from redirect pages, article text, inter-article links, and categories. We explore whether web links can replace a curated encyclopaedia, obtaining entity prior, name, context, and coherence models from a corpus of web pages with links to Wiki- pedia. Experiments compare web link models to Wikipedia models on well-known CoNLL and TAC data sets. Results show that using 34 million web links approaches Wikipedia performance. Combin- ing web link and Wikipedia models produces the best-known disambiguation accuracy of 88.7 on standard newswire test data. 1 Introduction Entity linking (EL) resolves mentions in text to their corresponding node in a knowledge base (KB), or NIL if the entity is not in the KB. Wikipedia and related semantic resources – Freebase, DBpedia, Yago2– have emerged as general repositories of no- table entities. The availability of Wikipedia, in par- ticular, has driven work on EL, knowledge base pop- ulation (KBP), and semantic search. This literature demonstrates that the rich structure of Wikipedia– redirect pages, article text, inter-article links, cat- egories – delivers disambiguation accuracy above 85% on newswire (He et al., 2013; Alhelbawy and Gaizauskas, 2014). But what disambiguation accu- racy can we expect in the absence of Wikipedia’s curated structure? Web links provide much of the same information as Wikipedia inter-article links: anchors are used to derive alternative names and conditional probabili- ties of entities given names; in-link counts are used to derive a simple entity popularity measure; the text surrounding a link is used to derive textual con- text models; and overlap of in-link sources is used to derive entity cooccurrence models. On the other hand, web links lack analogues of additional Wiki- pedia structure commonly used for disambiguation, e.g., categories, encyclopaedic descriptions. More- over, Wikipedia’s editors ensure a clean and correct knowledge source while web links are a potentially noisier annotation source. We explore linking with web links versus Wiki- pedia. Contributions include: (1) a new bench- mark linker that instantiates entity prior probabili- ties, entity given name probabilities, entity context models, and efficient entity coherence models from Wikipedia-derived data sets; (2) an alternative linker that derives the same model using only alternative names and web pages that link to Wikipedia; (3) de- tailed development experiments, including analysis and profiling of Web link data, and a comparison of link and Wikipedia-derived models. Results suggest that web link accuracy is at least 93% of a Wikipedia linker and that web links are complementary to Wikipedia, with the best scores coming from a combination. We argue that these re- sults motivate open publishing of enterprise author- ities and suggest that accumulating incoming links should be prioritised at least as highly as adding richer internal structure to an authority. 145 Transactions of the Association for Computational Linguistics, vol. 3, pp. 145–156, 2015. Action Editor: …

Content truncated to 500 words