摘要
针对预测型数据连续性较差、目标基数较大问题,文章提出了一种基于多维约束的多尺度预测型数据协同挖掘方法。通过参照数据的方式,进行目标数据实现尺度的划分,根据计算相似数据集间特征的频繁项,进行表达数据间的分布特性,建立祖先尺度数据集样本,计算其合集内各数据的相似性,利用JSC系数计算二者并集下相似度阈值较高的数据,实现数据协同挖掘。仿真实验证明,利用所提方法进行挖掘的效率更高、误判率较小、整体性能较为优异。
Aiming at the problems of poor continuity and a large target base of predictive data,this paper pro-poses a multi-scale collaborative data mining method based on multi-dimensional constraints.By re-ferring to data,the target data is divided into scales.According to the frequent term of the characteris-tics between similar data sets,the distribution characteristics between data are expressed,the ances-tor scale data set samples are established,the similarity of each data in the collection is calculated,and the data with high similarity threshold under the combination are calculated by JSC coefficient,and the data collaborative mining can be realized.The simulation results show that the proposed method has higher efficiency,lower error rate,and better overall performance.
作者
韩登宇
Dengyu Han(School of Health Science and Engineering,University of Shanghai for Science and Technology,Shanghai)
出处
《建模与仿真》
2025年第5期1176-1185,共10页
Modeling and Simulation
关键词
多维约束空间
权重值
离散型分布
相似度阈值
JSC系数
预测型数据
Multi-Dimensional Constraint Space
Weight Value
Discrete Distribution
Similarity Threshold
JSC Coefficient
Predictive Data
