Abstract: Ambiguity means inexactness. In case of data mining and natural language processing ambiguity refers to have more than one meaning of the same word. A term ‘Polysemy’ is also used for such a situation. This paper presents the process of ambiguity removal in data mining (so that the precise meaning of the word should be clear to the computer) and the problems with ambiguity removal. There are different types of ambiguity with a word. We will use probabilistic and statistical approach to remove the ambiguity attached with the word. Word Sense Disambiguation (WSD) is a must condition for human computer interaction otherwise without it the meaning of the word will not be clear to computer as user wants and this will create problems in further interaction between human and computer if user is dealing with a paragraph instead of a sentence. The probabilities for each word can be estimated from a tagged corpus.

Keywords: Ambiguity, POS Tag, Probability, Text mining