Modified PSO Based Feature Selection for Classification of Lung CT Images

Feature selection is an optimization problem in machine learning, which reduces the number of features, removes irrelevant, noisy and redundant data, and results in acceptable recognition accuracy. Feature selection is of great importance in pattern classification, medical data processing, machine learning, and data mining applications. In this paper, continuous particle swarm optimization (PSO) is used to implement a feature selection in wrapper based method, and the k-nearest neighbor classification serve as a fitness function of PSO for the classification problem. The PSO based feature selection method is applied to the features extracted from the Lung CT scan images. Experimental results show that modified PSO feature selection method simplifies features effectively and obtains a higher classification accuracy compared to the basic PSO Feature selection method. Keywords— Feature Selection, PSO, Population, Fitness function. INTRODUCTION Feature selection is the problem of selecting a subset of features without reducing the accuracy of representing the original set of features. Feature selection is used in many applications to remove irrelevant and redundant features where there are high dimensional datasets. These datasets may contain a high degree of irrelevant and redundant features that may decrease the performance of the classifiers. A Feature Selection algorithm explores the search space of different feature combinations to reduce the number of features and simultaneously optimize the classification performance. In Feature Selection, the size of the search space for n features is 2 [1] [3]. Feature selection is a multi-objective problem. It has two main objectives, which are to maximize the classification accuracy (minimize the classification error rate) and minimize the number of features. These two objectives are usually conflicting to each other and the optimal decision needs to be made in the presence of a trade-off between them. I. PARTICLE SWARM OPTIMIZATION PSO is an evolutionary computation technique proposed by Kennedy and Eberhart in 1995 [2] [4]. In PSO, a population, called a swarm, of candidate solutions are encoded as particles in the search space. PSO starts with the random initialization of a population of particles. The whole swarm move in the search space to search for the best solution by updating the position of each particle based on the experience of its own and its neighbouring particles [2] [3]. During movement, the current position of particle I is represented by a vector xi = (xi1, xi2, ..., xiD), where D is the dimensionality of the search space. The velocity of particle i is represented as vi = (vi1, vi2, ..., viD), which is limited by a predefined maximum velocity, vmax and vid [−vmax, vmax]. The best previous position of a particle is recorded as the personal best pbest and the best position obtained by the population thus far is called gbest. Based on pbest and gbest, PSO searches for the optimal solution by updating the velocity and the position of each particle according to the following equations: = + (1) = ∗ + ∗ ∗ ( − ) + ∗ ∗ − (2) where t denotes the t iteration, d denotes the d dimension in the search space D, w is inertia weight. c1 and c2 are acceleration constants. r1i and r2i are random values uniformly distributed in [0, 1]. pid and pgd represent the elements of pbest and gbest in the d dimension[5]. Figure 1: Structure of PSO based feature selection method Training Data Test Data PSO for Feature Selection Selected Features