Abstract: Entity Linking (EL) has received considerable attention in recent years. Given many name mentions in document, goal of EL is to predict their referent entities in knowledge base. Traditionally, there have been two distinct directions of EL research: one focusing on effects of mention's context compatibility, assuming that the referent entity of mention is reflected by its context; other dealing with effects of document's topic coherence, assuming that a mention's referent entity should be coherent with document's main topics. In this paper, we propose generative model -- called entity-topic model, to effectively join above two complementary directions together. By jointly modeling and exploiting context compatibility, topic coherence and correlation between them, our model can accurately link all mentions in document using both local information (including words and mentions in document) and global knowledge (including topic knowledge, entity context knowledge and entity name knowledge). Experimental results demonstrate effectiveness of proposed model.
Publication Year: 2012
Publication Date: 2012-07-12
Language: en
Type: article
Access and Citation
Cited By Count: 92
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot