Experimental building automatic Thesaurus By Using Data Mining

Abstract

In this paper, a new approach was presented, this approach building automatic thesaurus in the software engineering subject.
The proposed approach depend on the data Mining ( that refers to the overall process of discovering patterns or building models from a given data set) for indexing all the terms that founded into abstracts of the university thesis , this done by:
●Association rule: - this algorithm was used into data mining to find large item set and it is as a tool of discovery rule in our approach we used it to compute the frequency of the words into text and determined the keywords.
●Clustering techniques: it's used to classification of patterns
(observation, data items, or feature vectors into groups).
This new approach introduce automatic thesaurus that can be used as an effective tools for information retrieval.