基于机会认知的类脑智能数据挖掘机制  

Brain-Based Intelligent Data Mining Opportunities Cognitive Mechanisms

作　　者：宋小芹[1] 王莉丽[1] 张卫星[2] 

机构地区：[1]郑州大学西亚斯国际学院,河南新郑451150 [2]郑州大学软件与应用科技学院,河南郑州450002

出　　处：《计算机仿真》2016年第11期375-378,共4页Computer Simulation

基　　金：2016年度河南省高等学校重点科研项目(16A520094)

摘　　要：由于类脑智能数据具有协同多种不同的认知能力,对复杂环境具备极强的自适应能力,需要组建类脑智能数据的挖掘机制。但是采用当前的算法建立挖掘机制时,难以精确地消除数据中存在的冗余性,存在数据挖掘误差大的问题。为此提出基于机会认知的类脑智能数据挖掘机制。上述方法融合于时间粒度先对原始数据的时间序列进行分割,完成对数据的预处理,依据马氏距离来调整类脑智能数据变量之间的相关性,通过协方差矩阵得到其特征值和特征向量,将离散化的智能数据在粗糙集理论基础上进行属性的约简,在不损失原有类脑智能数据信息的基础上消除表中的冗余片段,从而组建了基于机会认知的类脑智能数据挖掘机制。实验结果表明,所提方法执行类脑智能数据挖掘效果较好,且挖掘效率较快。A mechanism of brain-like data mining based on the opportunity recognition is proposed. Firstly, the method segments the time series of original data integrated with time granularity and completes the data pretreatment. Then, the correlation between brain-like intelligent data variables is adjusted according to the Mahalanobis distance, and the eigenvalue and the eigenvector are obtained through the covariance matrix. Moreover, the attribute for the discretized intelligent data is reduced based on the rough set theory. Finally, the redundancy segment in the chart is eliminated based on the brain-like intelligent data information without loss, and the mechanism of brain-like intelligent data mining is built based on the opportunity recognition. The experimental results show that the method mentioned above has high mining efficiency and good effect of brain-like data mining.

关 键 词：机会认知 类脑智能 数据挖掘机制 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]