Based On The Nfcs-caf Data Mining Algorithm

Along with the fast and fierce development of the technology of computer and communication, the data become more and more. Thousands of databases are used in the fields of government, business, industry and so on. Say from a certain meaning, we are not short of the information, but were drowned by the information. Today, database system can efficiently realize the functions of record, modification, statistics, search, etc. But they can't discover the relations and rules in the data and forecast exactly the trend of the development of the future. So it causes phenomenon of the "banging of data but being short of knowledge". People need the new and more valid means to mine data.The data mining techniques produce and develop quickly under the application environment. They can converse the sea data into the useful knowledge automatically and intelligently. The discovering methods synthesize the research achievements in the fields including expert system, machine learning, statistics, pattern recognition, database, etc. They can resolve the problems in the information and in favor of data's conversion. They are regarded not only as the important task in the fields of database system and machine learning by researchers, but also as the important field which can bring large reword by businessmen.But in many cases, the traditional algorithms are hard to apply, or the application effects are not good. These cases include noisy data, redundant information, incomplete data, and sparse data in database. Neural networks can acquire knowledge by training set. They have many virtues such as parallelism, robustness, and anti-noise. And they can run efficiently even if the data are incomplete and noisy. So we advance the neural fuzzy cooperate system-combined activation function (NFCS-CAP) algorithms. It fuses strongpoint of both neural net and fuzzy system and adept to mine rules in database especially. The algorithm give over vice of the simple back propagation algorithm (SBP) that plunge the part extremes in some intricate nonlinear problems. Theoretical analyses and numerical simulation results have shown that the NFCS-CAF algorithm is feasible and effective. It provides a new way for studying the DM technique.