Word Sense Disambiguation for a domain-specific lexical sample task

Another limitation of already available supervised based word sense disambiguation (WSD) systems is that they only represent a word as a disorganize entity. However, representation of words using Vector Space Model can offer important and valuable information and hence improve overall accuracy. There are various methods that are already developed in the past. With some minor disadvantages, proposed relatedness system measures some properties for the upcoming generation of the Word Sense Disambiguation applications: sense inventory, domain understandability and universality. This paper investigates a novel approach in word sense disambiguation by exploiting every word in space and evaluates this approach on a domain-specific lexical sample task.