Know Your Graph. State-of-the-Art Knowledge-Based WSD

This paper introduces several improvements over the current state of the art in knowledge-based word sense disambiguation. Those innovations are the result of modifying and enriching a knowledge base created originally on the basis of WordNet. They reflect several separate but connected strategies: manipulating the shape and the content of the knowledge base, assigning weights over the relations in the knowledge base, and the addition of new relations to it. The main contribution of the paper is to demonstrate that the previously proposed knowledge bases organize linguistic and world knowledge suboptimally for the task of word sense disambiguation. In doing so, the paper also establishes a new state of the art for knowledge-based approaches. Its best models are competitive in the broader context of supervised systems as well.

[1]  Hwee Tou Ng,et al.  It Makes Sense: A Wide-Coverage Word Sense Disambiguation System for Free Text , 2010, ACL.

[2]  Martha Palmer,et al.  From TreeBank to PropBank , 2002, LREC.

[3]  Eneko Agirre,et al.  The risk of sub-optimal use of Open Source NLP Software: UKB is inadvertently state-of-the-art in knowledge-based WSD , 2018, ArXiv.

[4]  Rada Mihalcea,et al.  Unsupervised Large-Vocabulary Word Sense Disambiguation with Graph-based Algorithms for Sequence Data Labeling , 2005, HLT.

[5]  Kiril Ivanov Simov,et al.  Using Context Information for Knowledge-Based Word Sense Disambiguation , 2016, AIMSA.

[6]  Eneko Agirre,et al.  Using the Multilingual Central Repository for Graph-Based Word Sense Disambiguation , 2008, LREC.

[7]  Zhifang Sui,et al.  Incorporating Glosses into Neural Word Sense Disambiguation , 2018, ACL.

[8]  Zhifang Sui,et al.  Leveraging Gloss Knowledge in Neural Word Sense Disambiguation by Hierarchical Co-Attention , 2018, EMNLP.

[9]  Daniel Jurafsky,et al.  Learning to Merge Word Senses , 2007, EMNLP.

[10]  Roberto Navigli,et al.  Word Sense Disambiguation: A Unified Evaluation Framework and Empirical Comparison , 2017, EACL.

[11]  Hwee Tou Ng,et al.  One Million Sense-Tagged Instances for Word Sense Disambiguation and Induction , 2015, CoNLL.

[12]  Eneko Agirre,et al.  Random Walks and Neural Network Language Models on Knowledge Bases , 2015, NAACL.

[13]  Ido Dagan,et al.  context2vec: Learning Generic Context Embedding with Bidirectional LSTM , 2016, CoNLL.

[14]  Jeffrey Dean,et al.  Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[15]  John B. Lowe,et al.  The Berkeley FrameNet Project , 1998, ACL.

[16]  Egoitz Laparra,et al.  Predicate Matrix: extending SemLink through WordNet mappings , 2014, LREC.

[17]  Charles J. Fillmore,et al.  THE CASE FOR CASE. , 1967 .

[18]  Rajeev Motwani,et al.  The PageRank Citation Ranking : Bringing Order to the Web , 1999, WWW 1999.

[19]  Kiril Ivanov Simov,et al.  Comparison of Word Embeddings from Different Knowledge Graphs , 2017, LDK.

[20]  Michael E. Lesk,et al.  Automatic sense disambiguation using machine readable dictionaries: how to tell a pine cone from an ice cream cone , 1986, SIGDOC '86.

[21]  Eneko Agirre,et al.  Personalizing PageRank for Word Sense Disambiguation , 2009, EACL.

[22]  Roberto Navigli,et al.  Neural Sequence Learning Models for Word Sense Disambiguation , 2017, EMNLP.

[23]  Martha Palmer,et al.  Verbnet: a broad-coverage, comprehensive verb lexicon , 2005 .

[24]  Luciano Serafini,et al.  A novel Framenet-based resource for the semantic web , 2012, SAC '12.

[25]  Egoitz Laparra,et al.  Predicate Matrix: automatically extending the semantic interoperability between predicate resources , 2016, Lang. Resour. Evaluation.

[26]  George A. Miller,et al.  A Semantic Concordance , 1993, HLT.

[27]  Jeffrey Pennington,et al.  GloVe: Global Vectors for Word Representation , 2014, EMNLP.

[28]  Ignacio Iacobacci,et al.  Embeddings for Word Sense Disambiguation: An Evaluation Study , 2016, ACL.

[29]  Roberto Navigli,et al.  Entity Linking meets Word Sense Disambiguation: a Unified Approach , 2014, TACL.