A CHAIN DICTIONARY METHOD FOR WORD SENSE DISAMBIGUATION AND APPLICATIONS

Abstract. A large class of unsupervised algorithms for Word Sense Disam-biguation (WSD) is that of dictionary-based methods. Various algorithms have as the root Lesk’s algorithm, which exploits the sense definitions in the dictionary directly. Our approach uses the lexical base WordNet [3] for a new algorithm originated in Lesk’s, namely chain algorithm for disambigua-tion of all words (CHAD). We show how translation from a language into another one and also text entailment verification could be accomplished by this disambiguation. 1. The polysemy Word sense disambiguation is the process of identifying the correct sense of words in particular contexts. The solving of WSD seems to be AI complete ( that means its solution requires a solution to all the general AI problems of representing and reasoning about arbitrary) and it is one of the most important open problems in NLP [5],[6],[7], [10],[12],[13]. In the electronical on-line dictionary WordNet, the most well-developed and widely used lexical database for English, the poly-semy of different category of words is presented in order as: the highest for verbs, then for nouns, and the lowest for adjectives and adverbs. Usually, the process of disambiguation is realized for a single, target word. One would expect the words closest to the target word to be of greater semantical importance for it than the other words in the text. The context is hence a source of information to identify the meaning of the polysemous words. The contexts may be used in two ways: a) as bag of words, without consideration of relationships with the target word in terms of distance, grammatical relations, etc.; b) with relational information. The bag of words approach works better for nouns than verbs but is less effective than methods that take other relations in consideration. Studies about syntactic rela-tions determined some interesting conclusions: verbs derive more disambiguatio