摘要
针对当前数据挖掘的精确度及时效性较低的问题,提出基于尺度划分的多尺度数据挖掘算法.根据SVM(支持向量机)原理提取多元尺度数据集合共性特征,建立相应的数据层,根据样本评估模型对多尺度数据进行尺度分类,在此基础上利用信息熵原理求出多尺度数据映射函数,引入回归逻辑计算多尺度数据挖掘分类等级,结合多尺度划分数值上、下推处理原理求出复杂项集支持程度,完成多尺度数据挖掘.经实验分析,所提算法能有效提高数据挖掘的精确度和时效性,可广泛地应用于实际中.
Aiming at the problem of low accuracy and timeliness of the current data mining process,a multi-scale data mining algorithm is proposed based on scale division.Firstly,the common features of multi-scale data sets are extracted according to the principle of SVM(support vector machine)to establish the corresponding data layer.Secondly,the multi-scale data is classified according to the sample evaluation model,and then,the multi-scale data mapping function is obtained by the principle of information entropy.Finally,the classification level of multi-scale data mining is calculated by introducing the regression logic,and the degree of support for complex itemsets is obtained by combining the principle of multi-scale division of numerical up and down processing.In this way,the multi-scale data mining is completed.Experimental analysis shows that the proposed algorithm can effectively improve the accuracy and timeliness of data mining,and can be widely used in various fields.
作者
白玲玲
BAI Lingling(Educational Administration,Fuyang Party School of the CPC in Anhui Province,Fuyang Anhui 236000)
出处
《宁夏师范学院学报》
2020年第7期65-72,共8页
Journal of Ningxia Normal University
关键词
尺度划分
数据挖掘
特征样本
映射函数
信息熵理论
Scale division
Data mining
Feature samples
Mapping function
Information entropy theory
