Research And Application Of Artificial Neural Network In Data Mining

Along with the application of computer system in different fields,the data are continuously stored in the database,internet and so on.Among them,people expect to find out potential knowledge that will help them make decisions.The emergence and development of data mining are just based on this expectation.As a midpoint of different subject studies,data mining involves a great deal of subjects such as database,statistic,machine learning,artificial intelligence,high performance computing,pattern recognition and data visualization etc.Among them,artificial neural network is widely used due to its ability of inherent non-linear processing, adaptive learning and high fault-tolerance.In the paper,we study that artificial neural network was widely used in data mining.We use the fitness based on information of mean square deviation and error uniformity as a compound fitness training index of PSO,and an improved PSO by use the simples method(SM) in order to improve the generalization ability and the optimization ability of neural networks.Finally,a neural network with compound fitness and SMPSO algorithm is constructed.And we use this neural network in an IRIS data mining program.The result show that the system precision and the noise contain capability is greatly improved.For the internet can search a lot of information,some of bad information such as sex violence and so on can spread widely and make bad influence to our society.How to filter the bad information and build up the health,safety information environment has become a research focus of the internet development in the future.Building a module that can judge some of bad information automatically by the SMPSO-ANN which discussed in this article and filter these based on an open source web filter project Dansguardian is a milestone from abstract research to actual application.We made it by use the C language and it has some of innovate value.