Mixing Context Granularities for Improved Entity Linking on Question Answering Data across Entity Categories

Proceedings of the 7th Joint Conference on Lexical and Computational Semantics (*SEM), pages 65–75 New Orleans, June 5-6, 2018. c⃝2018 Association for Computational Linguistics Mixing Context Granularities for Improved Entity Linking on Question Answering Data across Entity Categories Daniil Sorokin and Iryna Gurevych Ubiquitous Knowledge Processing Lab (UKP) and Research Training Group AIPHES Department of Computer Science, Technische Universit¨at Darmstadt www.ukp.tu-darmstadt.de {sorokin|gurevych}@ukp.informatik.tu-darmstadt.de Abstract The first stage of every knowledge base question answering approach is to link en- tities in the input question. We investigate entity linking in the context of a question answering task and present a jointly opti- mized neural architecture for entity men- tion detection and entity disambiguation that models the surrounding context on dif- ferent levels of granularity. We use the Wikidata knowledge base and available question answering datasets to create benchmarks for entity linking on question answering data. Our approach outperforms the previous state-of-the-art system on this data, resulting in an average 8% improvement of the final score. We fur- ther demonstrate that our model delivers a strong performance across different entity categories. 1 Introduction Knowledge base question answering (QA) requires a precise modeling of the question semantics through the entities and relations available in the knowledge base (KB) in order to retrieve the cor- rect answer. The first stage for every QA approach is entity linking (EL), that is the identification of entity mentions in the question and linking them to entities in KB. In Figure 1, two entity mentions are detected and linked to the knowledge base ref- erents. This step is crucial for QA since the correct answer must be connected via some path over KB to the entities mentioned in the question. The state-of-the-art QA systems usually rely on off-the-shelf EL systems to extract entities from the question (Yih et al., 2015). Multiple EL sys- tems are freely available and can be readily applied what are taylor swift’s albums ? Taylor Swift Q462 album Q24951125 Red, 1989, etc. PERFORMER INSTANCE OF Figure 1: An example question from a QA dataset that shows the correct entity mentions and their relationship with the correct answer to the question, Qxxx stands for a knowledge base identifier for question answering (e.g. DBPedia Spotlight1, AIDA2). However, these systems have certain draw- backs in the QA setting: they are targeted at long well-formed documents, such as news texts, and are less suited for typically short and noisy ques- tion data. Other EL systems focus on noisy data (e.g. S-MART, Yang and Chang, 2015), but are not openly available and hence limited in their us- age and application. Multiple error analyses of QA systems point to entity linking as a major external source of error (Berant and Liang, 2014; Reddy et al., 2014; Yih et al., 2015). The QA datasets are normally collected from the web and contain very noisy and diverse data (Be- rant et al., 2013), which poses a number of chal- lenges for EL. First, many common features used in EL systems, such as capitalization, are …

Content truncated to 500 words