Title: HIT Approaches to Entity Linking at TAC 2012
Abstract: This paper describes the system of HIT at the 2011 Text Analysis Conference (TAC) Knowledge Base Population (KBP) track English Entity Linking task. In this task, a system is required to link a name string in a given contextual document to its referent entity in an external knowledge base. The HIT system makes the linkage by using the relatedness score between the context and the model of the entity. In this system, an entity is modeled based on its information from Wikipedia. The relatedness score contains three parts: the popularity of the entity along with the query name, the context language model score which is generated by the entity model and the occurrence of the aliases of the entity. The evaluation result shows the system performance is comparable with the median performance of the participating systems.
Publication Year: 2012
Publication Date: 2012-01-01
Language: en
Type: article
Access and Citation
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot