Joint Entity Recognition and Disambiguation

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 879–888, Lisbon, Portugal, 17-21 September 2015. c⃝2015 Association for Computational Linguistics. Joint Named Entity Recognition and Disambiguation Gang Luo1, Xiaojiang Huang2, Chin-Yew Lin2, Zaiqing Nie2 1Microsoft, California, USA 2Microsoft Research, Beijing, China {gluo, xiaojih, cyl, znie}@microsoft.com Abstract Extracting named entities in text and link- ing extracted names to a given knowledge base are fundamental tasks in applications for text understanding. Existing systems typically run a named entity recognition (NER) model to extract entity names first, then run an entity linking model to link ex- tracted names to a knowledge base. NER and linking models are usually trained sep- arately, and the mutual dependency be- tween the two tasks is ignored. We pro- pose JERL, Joint Entity Recognition and Linking, to jointly model NER and link- ing tasks and capture the mutual depen- dency between them. It allows the in- formation from each task to improve the performance of the other. To the best of our knowledge, JERL is the first model to jointly optimize NER and linking tasks to- gether completely. In experiments on the CoNLL’03/AIDA data set, JERL outper- forms state-of-art NER and linking sys- tems, and we find improvements of 0.4% absolute F1 for NER on CoNLL’03, and 0.36% absolute precision@1 for linking on AIDA. 1 Introduction In applications of complex Natural Language Pro- cessing tasks, such as automatic knowledge base construction, entity summarization, and question answering systems, it is essential to first have high quality systems for lower level tasks, such as part- of-speech (POS) tagging, chunking, named en- tity recognition (NER), entity linking, and parsing among others. These lower level tasks are usually decoupled and optimized separately to keep the system tractable. The disadvantage of the decou- pled approach is that each lower level task is not aware of other tasks and thus not able to leverage information provided by others to improve perfor- mance. What is more, there is no guarantee that their outputs will be consistent. This paper addresses the problem by building a joint model for Entity Recognition and Disam- biguation (ERD). The goal of ERD is to extract named entities in text and link extracted names to a knowledge base, usually Wikipedia or Freebase. ERD is closely related to NER and linking tasks. NER aims to identify named entities in text and classify mentions into predefined categories such as persons, organizations, locations, etc. Given a mention and context as input, entity linking con- nects the mention to a referent entity in a knowl- edge base. Existing ERD systems typically run a NER to extract entity mentions first, then run an entity linking model to link mentions to a knowledge base. Such a decoupled approach makes the sys- tem tractable, and both NER and linking models can be optimized separately. The disadvantages are also obvious: 1) errors caused by NER will be propagated to linking and are not recoverable 2) NER can not benefit from information available used in entity …

Content truncated to 500 words