Entity Linking for Tweets

Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics, pages 1304–1311, Sofia, Bulgaria, August 4-9 2013. c⃝2013 Association for Computational Linguistics Entity Linking for Tweets Xiaohua Liu†, Yitong Li‡, Haocheng Wu♯, Ming Zhou†, Furu Wei†, Yi Lu§ †Microsoft Research Asia, Beijing, 100190, China ‡School of Electronic and Information Engineering Beihang University, Beijing, 100191, China ♯University of Science and Technology of China No. 96, Jinzhai Road, Hefei, Anhui, China §School of Computer Science and Technology Harbin Institute of Technology, Harbin, 150001, China †{xiaoliu, mingzhou, fuwei}@microsoft.com ‡tong91222@126.com ♯v-haowu@microsoft.com §v-y@microsoft.com Abstract We study the task of entity linking for tweets, which tries to associate each mention in a tweet with a knowledge base entry. Two main challenges of this task are the dearth of information in a single tweet and the rich entity mention variations. aTo address these challenges, we propose collective inference method that simultaneously resolves a set of mentions. Particularly, our model integrates three kinds of similarities, i.e., mention-entry similarity, entry-entry similarity, and mention-mention similarity, to enrich the context for entity linking, and to address irregular mentions that are not covered by the entity-variation dictionary. We evaluate our method on a publicly available data set and demonstrate the effectiveness of our method. 1 Introduction Twitter is a widely used social networking service. With millions of active users and hundreds of millions of new published tweets every day1, it has become a popular platform to capture and transmit the human experiences of the moment. Many tweet related researches are inspired, from named entity recognition (Liu et al., 2012), topic detection (Mathioudakis and Koudas, 2010), clustering (Rosa et al., 2010), to event extraction (Grinev et al., 2009). In this work, we study the entity linking task for tweets, which maps each entity mention in a tweet to a unique entity, i.e., an entry ID of a knowledge base like Wikipedia. Entity 1http://siteanalytics.compete.com/twitter.com/ linking task is generally considered as a bridge between unstructured text and structured machine- readable knowledge base, and represents a critical role in machine reading program (Singh et al., 2011). Entity linking for tweets is particularly meaningful, considering that tweets are often hard to read owing to its informal written style and length limitation of 140 characters. Current entity linking methods are built on top of a large scale knowledge base such as Wikipedia. A knowledge base consists of a set of entities, and each entity can have a variation list2. To decide which entity should be mapped, they may compute: 1) the similarity between the context of a mention, e.g., a text window around the mention, and the content of an entity, e.g., the entity page of Wikipedia (Mihalcea and Csomai, 2007; Han and Zhao, 2009); 2) the coherence among the mapped entities for a set of related mentions, e.g, multiple mentions in a document (Milne and Witten, 2008; Kulkarni et al., 2009; Han and Zhao, 2010; Han et al., 2011). Tweets pose special challenges to entity linking. First, a tweet is often too concise and too …

Content truncated to 500 words