数学建模中的高维数据挖掘技术优化研究

高维数据挖掘由于特征空间占用开销较大,挖掘的复杂度较高,挖掘精度不高,为了提高对高维数据挖掘的准确性能,提出一种基于相空间重构和K-L变换特征压缩的高维数据挖掘数学建模方法。采用集成学习技术,对高维数据信息流进行相空间重构处理,考虑类间的数据不平衡性,求得高维数据的关联维特征参量,根据数据的链距离进行稀疏性融合,计算高维数据流模型的最大Lyapunove指数谱,根据谱分析方法实现数据聚类,对聚类后的数据采用K-L特征压缩方法进行降维处理,降低数据挖掘的内存及计算开销。仿真结果表明,采用该方法进行高维数据挖掘,数据挖掘的准确概率较高,占用内存消耗较少,计算开销较小。

Research on Optimization of high dimensional data mining in mathematical modeling

High dimensional data mining due to the characteristics of the space occupied large overhead mining, high complexity, mining precision is not good, in order to improve the accuracy of performance on high dimensional data mining, this paper brings forward a mining method of mathematical modeling of phase space reconstruction and K-L transform features of high dimensional data based on compression. The ensemble learning technique to reconstruct the phase space of high dimensional data flow, considering the inter class data imbalance, the correlation dimension of the characteristic parameters of high dimensional data, according to the chain distance data sparsity fusion, maximum Lyapunove computation of high dimensional data stream model refers to the number of spectra, the spectral analysis method of data after clustering, clustering of data using K-L feature dimension compression method, reduce the memory and computation overhead of data mining. The simulation results show that the method has high accuracy, less memory consumption and less computation cost.