4.0 DOCUMENT RETRIEVAL
In the proposed model for domain knowledge, we generated a matrix defining relationship between concepts and clusters i.e. we may find out clusters that are the best representative of individual concepts. The concept vector is compared to every document vector within respective optimal cluster. This identifies the document coming closest to the query and determines the rank of closeness of fit to the others. Similarity function be- tween the concept vector and the document vector [24] in the optimal cluster is used for retrieving the documents Suppose the document Di is represented as a vector of dimension t,
and concept Cj as a t-dimension vector of the form
The similarity between the two items can be obtained as the inner product between corresponding weighted term vector as follows [23]