This paper proposes a novel unsupervised model for all-words word sense disambiguation (WSD) to cope with the enormous number of sense classes inherent in the task. The proposed model is a hierarchical Bayesian model that incorporates two types of soft constraints and infers natural correspondence between unlabeled word occurrences and numerous senses: 1) senses of word instances follow the prior distribution of each word-type, 2) senses in a context follow the extrapolation from other words’ senses in similar context. Experimental results applied to SemEval dataset confirmed the advantages of our hierarchical model.

This paper proposes a novel unsupervised method employing large amount of unlabeled text cor- pora for all-words word sense disambiguation (WSD), which requires to discriminate huge variety of senses, thus unsupervised methods are desired to avoid constructing costly sense-labeled corpora. Given unlabeled corpora and a dictionary, the proposed method bases on the coherent correspondences between word con- texts and word senses, and finds the all-words’ senses that maximize mapping density in context-to-sense product metric space. Experimental results confirmed the efficacy of our unsupervised method by showing the reliability of disambiguation if sufficient variations of word-types are provided in similar context.

This paper proposes a novel smoothing model with a combinatorial optimization scheme for all-words word sense disambiguation from untagged corpora. By generalizing discrete senses to a continuum, we introduce a smoothing in context-sense space to cope with data-sparsity resulting from a large variety of linguistic context and sense, as well as to exploit sense-interdependency among the words in the same text string. Through the smoothing, all the optimal senses are obtained at one time under maximum marginal likelihood criterion, by competitive probabilistic kernels made to reinforce one another among nearby words, and to suppress conflicting sense hypotheses within the same word. Experimental results confirmed the superiority of the proposed method over conventional ones by showing the better performances beyond most-frequent-sense base-line performance where none of SemEval-2 unsupervised systems reached.