Feature Diminution by Ant Colonized Relative Reduct Algorithm for improving the Success Rate for IVF Treatment

Abstract

Infertility is the most common problem faced by today’s generation. The factors like environment, genetic or personal characteristics are responsible for these problems. Different infertility treatments like IVF, IUI etc are used to treat those infertile people. But the cost and emotions beyond each and every cycle of IVF treatment is very high and also the success rate differs from person to person. So, there is a need to find a system which would predict the outcome of IVF to motivate the people both in psychologically and financially. Many Data Mining techniques are applied to predict the outcome of the IVF treatment. Reducing the unwanted features which affects the quality of result is one of the significant tasks in Data Mining. This paper proposes a hybrid algorithm named Ant Colonized Relative Reduct Algorithm (ACRRA) which combines the core features of Ant Colony Optimization Algorithm and Relative Reduct Theory for Feature Reduction. In this work, the proposed Algorithm is compared with the existing related algorithms. It is evident from the results that the proposed algorithm achieved its target of reducing the features to minimum numbers without compromising the core knowledge of the system to estimate the success rate.