多尺度混合算法在智慧能源需求数据挖掘中的应用

Application of multi-scale hybrid algorithm in intelligent energy demand data mining

针对传统算法的缺陷,提出了基于多尺度混合算法的数据挖掘使用。首先,基于概念分层理论提出数据尺度划分。之后,以多尺度理论研究的重点,对多尺度数据挖掘实质进行分析,基于多尺度数据理论研究提出多尺度混合算法的数据挖掘技术。此算法通过采样理论实现数据集挖掘结果频繁项集的处理,实现多尺度数据间知识的推导。最后,对算法进行实验与分析。通过算法测试表示,此算法的精准度、覆盖度比较高,支持度估计误差低,有效且可行。

Aiming at the defect of traditional algorithm,a data mining method based on multi-scale hybrid algorithm is proposed. Firstly,based on the theory of concept stratification,the data scale division is proposed. After that,the essence of multi-scale data mining is analyzed with the emphasis of multi-scale theory research. Based on the multi-scale data theory research,the multi-scale hybrid algorithm data mining technology is proposed. In this algorithm,the frequent itemsets of data mining results are processed by sampling theory,and the knowledge among multi-scale data is derived. Finally,the algorithm is tested and analyzed. The algorithm test shows that the accuracy and coverage of this algorithm is relatively high,and the support estimation error is low,which is effective and feasible.