基于数据挖掘算法的决策支持系统的优化研究

Research on Optimization of Decision Support System Based on Data Mining Algorithm

将信息熵对信息和数据的不确定性分析来度量数据所带来的不确定性程度,利用数据挖掘算法中的蚁群聚类算法,结合信息熵理论对网络客户数据进行分析,其中,信息熵理论中的不确定性分析,可以较好的帮助聚类数据对象,数据在此基础上进行再重组,其结果可进一步提高决策的有效性。

In order to optimize data mining algorithms in decision support system(DSS),this paper proposes a SRF algorithm based on random forest algorithm,which is suitable for distributed storage of decision support system,and has better classification results under parallel computing. The classification accuracy of the improved SRF algorithm and the speed of parallel processing are simulated in this paper.Simulation results demonstrate that the data mining and data warehouse via SRF algorithm can make up for shortcomings of the original decision support system,and also can take advantage of the current system of database resources effectively by adjusting the mutual relationship among system data, research models,experimental methods. Therefore,the integrated performance of system will be greatly improved.