New Unification Matching Scheme for efficient information retrieval using Genetic Algorithm

This article presents a new Unification Matching Scheme (UMS) for information retrieval using the genetic algorithm. The selection of appropriate matching functions contributes to the performance of the information retrieval system. The proposed UMS executes the Unification function on three classical matching functions for different threshold values. The main objective is to utilize all the base functions to increase the relevancy of a users query with the data objects. The best results from each matching function define the new generation on, which the other matching functions are applied. The results from each generation are optimized using the Genetic Algorithm. The working of UMS is compared with individual classical matching functions. A significant improvement is seen in the experimental results in terms of precision and recall. The performance increased/increases gradually, in each generation thereby, producing the relevant results.