Neural Collective Entity Linking

Proceedings of the 27th International Conference on Computational Linguistics, pages 675–686 Santa Fe, New Mexico, USA, August 20-26, 2018. 675 Neural Collective Entity Linking Yixin Cao, Lei Hou∗, Juanzi Li, Zhiyuan Liu Dept. of Computer Science and Technology, Tsinghua University, China 100084 {caoyixin2011,greener2009,lijuanzi2008}@gmail.com liuzy@tsinghua.edu.cn Abstract Entity Linking aims to link entity mentions in texts to knowledge bases, and neural models have achieved recent success in this task. However, most existing methods rely on local contexts to re- solve entities independently, which may usually fail due to the data sparsity of local information. To address this issue, we propose a novel neural model for collective entity linking, named as NCEL. NCEL applies Graph Convolutional Network to integrate both local contextual features and global coherence information for entity linking. To improve the computation efficiency, we approximately perform graph convolution on a subgraph of adjacent entity mentions instead of those in the entire text. We further introduce an attention scheme to improve the robustness of NCEL to data noise and train the model on Wikipedia hyperlinks to avoid overfitting and domain bias. In experiments, we evaluate NCEL on five publicly available datasets to verify the linking performance as well as generalization ability. We also conduct an extensive analysis of time com- plexity, the impact of key modules, and qualitative results, which demonstrate the effectiveness and efficiency of our proposed method. 1 Introduction Entity linking (EL), mapping entity mentions in texts to a given knowledge base (KB), serves as a fun- damental role in many fields, such as question answering (Zhang et al., 2016), semantic search (Blanco et al., 2015), and information extraction (Ji et al., 2015; Ji et al., 2016). However, this task is non-trivial because entity mentions are usually ambiguous. As shown in Figure 1, the mention England refers to three entities in KB, and an entity linking system should be capable of identifying the correct entity as England cricket team rather than England and England national football team. Entity linking is typically broken down into two main phases: (i) candidate generation obtains a set of referent entities in KB for each mention, and (ii) named entity disambiguation selects the possible candidate entity by solving a ranking problem. The key challenge lies in the ranking model that computes the relevance between candidates and the corresponding mentions based on the information both in texts and KBs (Nguyen et al., 2016). In terms of the features used for ranking, we classify existing EL models into two groups: local models to resolve mentions independently relying on textual context information from the surrounding words (Chen and Ji, 2011; Chisholm and Hachey, 2015; Lazic et al., 2015; Yamada et al., 2016), and global (collective) models, which are the main focus of this paper, that encourage the target entities of all mentions in a document to be topically coherent (Han et al., 2011; Cassidy et al., 2012; He et al., 2013b; Cheng and Roth, 2013; Durrett and Klein, 2014; Huang et al., 2014). Global models usually build an …

Content truncated to 500 words