Abstract:Word sense disambiguation (WSD) is a fundamental natural language processing task. Unsupervised knowledge-based WSD only relies on a lexical knowledge base as the sense inventory and has wider practical use than supervised WSD that requires a mass of sense-annotated data. HowNet is the most widely used lexical knowledge base in Chinese WSD. Because of its uniqueness, however, most of existing unsupervised WSD methods cannot work for HowNet-based WSD, and the tailor-made methods have not obtained satisfying results. In this paper, we propose a new unsupervised method for HowNet-based Chinese WSD, which exploits the masked language model task of pre-trained language models. In experiments, considering existing evaluation dataset is small and out-of-date, we build a new and larger HowNet-based WSD dataset. Experimental results demonstrate that our model achieves significantly better performance than all the baseline methods. All the code and data of this paper are available at https://github.com/thunlp/SememeWSD.

Research on Hownet-Based Chinese Word Lexical Semantic Similarity Measurement

Quantifying Semantic Similarity of Chinese Words from HowNet

A New Hypred Improved Method for Measuring Concept Semantic Similarity in WordNet.

An New Approach Measuring Semantic Similarity in Hownet 2000

Word semantic similarity computation based on integrating HowNet and search engines

Measuring Chinese-English Cross-Lingual Word Similarity With Hownet And Parallel Corpus

A Novel Comprehensive Approach for Estimating Concept Semantic Similarity in WordNet

Word Similarity Computation Based on Wordnet and Hownet

Computing semantic similarity using Machine-Readable Dictionary and HowNet sememe

An Approach of Semantic Similarity by Combining HowNet and Cilin.

Chinese Word Similarity Computing Based on Semantic Tree

Semantic Field Construction Based on HowNet

Researches of Chinese Sentence Similarity Based on HowNet

Semantic Similarity Computation Based on Multi-feature Combination Using HowNet.

Semantic Similarity Calculation Based on Sememe Set

Application-Oriented Comparison and Evaluation of Six Semantic Similarity Measures Based on Wordnet

Word similarity measurement based on basic lexical set

Try to Substitute: an Unsupervised Chinese Word Sense Disambiguation Method Based on HowNet.

A Similarity Algorithm Based on the Generality and Individuality of Words

Semantic Orientation Computing Based on HowNet

The Automatic Construction of Lexical Semantic Relationship Graph Based on HowNet