The Particle Swarm Optimization And Its Application

Particle Swarm Optimization (PSO) is a swarm based intelligence optimization method, it solves optimization problems by simulating the social behavior of bird flocks. PSO has been widely applied and developed in the field of electrical, communication, control and so on. It takes advantage of colony to find new avenue for the solution of complex problems. Therefore, to study and master the characteristics and rule of PSO is a significant task.As humans, it is easy to recognize letters, numbers, voices, etc. However, making a computer solve these types of problems is a very difficult task. Pattern recognition is the science with the objective to classify objects into different categories and classes. It is a fundamental component of artificial intelligence and computer vision. Clustering analysis is an important branch of pattern recognition, it is an unsupervised pattern recognition method, and was widely used in many fields. With the development of computer technology, traditional clustering technology in combination with modern optimization methods becomes available.In recent decades, in the face of massive data in the information age, data mining is arose and rapidly developed. As one of the most important models of data mining, this technology by which the knowledge discovered can be used to offer decision support, has the most significant application value. This thesis researches on the basic theory of discrete particle swarm optimization, and then presents an improved algorithm, which is introduced into data mining.Introduces a new algorithm based on PSO to the field of clustering. The algorithm is successfully applied to clustering problems including Iris, etc. The experimental results show that this algorithm can obtain good clustering results compared with K-means algorithm, it can improve the clustering accuracy with lower error rate. Then a new technology , combined with genetic algorithms, is presented based on particle swarm optimizer, including particle coding, population initializing, fitness computing and position updating. At the same time, multi-populations and selection operator of genetic algorithm are introduced into it, which forms a new algorithm, naming Multi-Populations Hybrid Particle Swarm Optimization. Then, an experiment is tried out on a MS database–mushroom, and experimental results show that particle swarm optimizer has better quality and performance compared with genetic algorithm.