Word Sense Disambiguation using WordNet Relations and Parallel Corpora

Word Sense Disambiguation is one of the most important challenge in computational linguistics. It is often described as "AI-complete" and is the most critical issue in natural language processing. Although it has been addressed by many researchers, no satisfactory results are reported. Rule based systems alone cannot handle this issue due to ambiguous nature of the natural language. Knowledge-based systems are therefore essential to find the intended sense of a word form. Machine readable dictionaries have been widely used in word sense disambiguation. The problem with this approach is that the dictionary entries for the target words are very short. WordNet is the most developed and widely used lexical database for English. The entries are always updated and many tools are available to access the database on all sorts of platforms. The WordNet database can be conveted in MySQL format and we have modified it as per our requirement. Sense's definitions of the specific word, "Synset" definitions, the "Hypernymy" relation, and definitions of the context features (words in the same sentence) are retrieved from the WordNet database and used as an input of our Disambiguation algorithm.