Improvised Gene Selection Using Particle Swarm Optimization With Decision Tree As Classifier

Pranav Teja Garikapati , Naveen Kumar Penki, Sashank Gogineni

Innovation has spread its foundations profound into the lives of a cutting-edge man, and the essential perspective to get the life going, health care is not an exemption. In spite of the fact that numerous researchers are contributing for inquiring about in the field of Bio-Technology, not many contributed for identification of cancer by selecting prominent genes from a microarray gene data set. Here, we intend to perform gene selection using Particle Swarm Optimization in a novel method, by combining it with other existing classification algorithm to arrive at a better accuracy in cancer identification, as compared to the already existing one. In this study, we implemented a novel method for medical problem, it is the integration of particle swarm optimization (PSO) and decision tree (C4.5) named PSO + C4.5 algorithm. To evaluate the effectiveness of PSO + C4.5 algorithm, it is implemented on cancer data set of life sciences obtained from UCI machine learning databases. The fitness value is calculated using C4.5 and this improves the PSO algorithm’s efficiency in obtaining the characteristic genes.