Feature Selection and Classification for Microarray Data Using ACO-FLANN Framework

Classification is a data mining technique used to predict group membership for data instances. The main objective of a classifier is to discover the hidden class level of the unknown data. The classifier performance depends upon the data size, number of classes, and dimension of feature spaces. The classifier accuracy has been improved by applying optimization techniques. There are many optimization techniques that are developed for this purpose such as Particle Swarm Optimization (PSO), ACO, ABC, DE, MLP, FLANN, PSO-FLANN, etc. In this proposed work, a new model is proposed for the classification of microarray data. Artificial Neural Network(ANN) uses Ant Colony Optimization(ACO) to tune the parameters of ANN. Principal Component Analysis (PCA) is used for dimensionality reduction. Here in the first step, the reduced dataset is optimized by using Ant Colony Optimization(ACO) and after that, in the second step, the optimized dataset is trained to Functional Link Artificial Neural Network (FLANN). This model is called ACO-FLANN. The proposed model(ACO-FLANN) has been compared with PSO-FLANN. The simulation shows that the proposed classification technique is superior and faster than PSO-FLANN.