多媒体智能教学系统中特定数据挖掘方法研究

摘要

The multimedia teaching system in particular key data are accurate, could improve the compatibility and multimedia intelligent teaching system of information data access capabilities. The traditional method using empirical mode decomposition method for data mining, when the scale enlargement of multimedia data intelligent teaching system, the improvement of information alignment, the accuracy of data mining. In this paper, a criterion based on affine transformation and specific data stream data mining algorithm of phase space reconstruction, multimedia intelligent teaching system established by the structure model of the distribution of data of the data by dimensional affine transform information fusion processing, phase space reconstruction was carried out on the merged data traffic, in higher dimensional phase space reconstruction to extract the smart multimedia teaching systems of higher order moments of the specific data in feature, realize the characteristic data of accurate mining. The simulation results show that by using this method to the accuracy of the data mining to identify probability is higher, anti-interference performance is stronger.%对多媒体教学系统中特定关键数据进行准确挖掘，可以提高多媒体智能教学系统的信息兼容和数据访问能力。传统方法采用经验模态特征分解方法进行数据挖掘，当多媒体智能教学系统数据规模的扩大、信息融合度的提高时，数据挖掘的准确度下降。提出一种基于尺度仿射变换和特定数据信息流相空间重构的数据挖掘算法，首先建立多媒体智能教学系统的数据分布结构模型，采用尺度仿射变换对数据进行信息融合处理，对融合后的数据信息流进行相空间重构，在重构的高维相空间中提取多媒体智能教学系统中特定数据的高阶矩特征，实现对特征数据的准确挖掘。仿真结果表明，采用该方法进行数据挖掘的准确识别概率较高，抗干扰性能较强。