Main Article Content
The two notable evolutionary computational and swarm based approach used for combinatorial optimization problems which have a discrete pursuit space are Ant Miner algorithm (AM) and genetic algorithm (GA). The feature selection process selects a feature subset and then processes the data with chosen features to learn the algorithm. A classification model gets executed in the prediction stage using the last chosen feature. The hybrid algorithm (AM-GA) combines the distinctive features of ants and genetic approach to optimize the attribute reduction. By hybridization, the space intricacy is decreased by eliminating the stagnant behavior of ants and the time complexity is reduced by the global search mechanism adopted by the genetic algorithm. The performance of the proposed approach has been evaluated on the PIMA Indian Type II diabetic datasets taken from the UCI machine learning repository. The experimental results prove that the proposed approach has selected the best possible features for achieving the highest rate of classification accuracy.
This work is licensed under a Creative Commons Attribution 4.0 International License.