A Model for Parallel Data Mining Based on Multi-Agent

With the increasing number of data mining process, frequency of affairs distribution and results collection will increase in exponent, thus resulting in system overload. In order to improve the efficiency of data mining, this paper designed a parallel data mining model for centralized dataset only. The intelligent algorithm of the model reduces the system operation cost significantly and balances system load reasonably as well. The model system structure and the basic services each module provided are listed. The experiment proves that the model is suitable for the large scale, high precision and multi-dimensional data mining, it can also notably reduce the running time of mining process.