度量相似性数学建模_数学建模中的高维数据挖掘技术优化研究-Research on Optimization of high dimensional data mining in mathematica...

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.