基于改进密度峰值聚类的企业数据管理研究  

Research on Enterprise Data Management Based on Improved Density Peak Clustering

作　　者：黄铿[1] 陈易平 李庆珍[3] HUANG Keng;CHEN Yi-ping;LI Qing-zhen(School of Economics and Management,Guangzhou Nanyang Polytechnic College,Guangzhou 510900,Guangdong,China;School of Computer Science and Engineering,Guangzhou Institute of Technology,Guangzhou 510540,Guangdong,China;Institute of data rule of law,China University of political science and law,Beijing 102249,China)

机构地区：[1]广州南洋理工职业学院经济管理学院,广东广州510900 [2]广州理工学院计算机科学与工程学院,广东广州510540 [3]中国政法大学数据法治研究院,北京102249

出　　处：《贵阳学院学报（自然科学版）》2023年第4期61-66,共6页Journal of Guiyang University：Natural Sciences

基　　金：2019年广东省教育厅特色创新项目“基于机器人视觉的目标跟踪技术应用研究”(编目编号:2019KTSCX243)。

摘　　要：为提高企业数据管理的有效性,采用密度峰值聚类(DPC)算法用于企业数据聚类,将数据有序化和类别化。首先,对企业数据样本进行清洗并按权重对指标进行特征提取,并建立DPC的企业数据分类模型。然后,在DPC的聚类簇中心选择时,为防止该值设置不合理使得密度值和距离值偏移较大而影响聚类精度,采用鲸群优化算法对距离阈值优化求解。此外,为增强搜索精度,将鲸群坐标用量子比特表示,从而实现了量子鲸鱼优化算法(QWOA)。最后,采用QWOA优化得到的距离阈值进行DPC聚类,获得企业数据分类结果。实验结果表明,合理设置鲸群规模和选择概率,QWOA-DPC算法能够获得较高的分类精度,通过对3类不同行业的QWOA-DPC企业数据进行分析,均得到了较高的分类性能,为企业数据管理提供了有效的策略支持。Because of the large amount of enterprise data,strong data heterogeneity and high dimensionality,and the strong correlation between upstream and downstream relational data of enterprise operation,enterprise data management is becoming more and more complicated.In order to improve the effectiveness of enterprise data management,the density peak clustering(DPC)algorithm is used for enterprise data clustering,which makes the data orderly and classified.Firstly,the enterprise data samples are cleaned and the characteristics of indicators are extracted according to the weights,and the enterprise data classification model of DPC is established.Then,in the selection of DPC cluster center,in order to prevent the unreasonable setting of this value from causing the density value and distance value to deviate greatly and affecting the clustering accuracy,the whale swarm optimization algorithm is used to optimize the distance threshold.In addition,in order to enhance the search accuracy,the whale group coordinates are represented by quantum bits,and thus the Quantum Whale Optimization Algorithm(QWOA)is implemented.Finally,the distance threshold obtained by QWOA optimization is used for DPC clustering to obtain the classification results of enterprise data.The experimental results show that the QWOA-DPC algorithm can achieve high classification accuracy by reasonably setting the whale population size and selection probability.Through the analysis of three different industries'QWOA-DPC enterprise data,high classification performance is obtained,which provides effective strategic support for enterprise data management.

关 键 词：企业数据管理 密度峰值聚类 量子比特 鲸群优化算法 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术] TP311.13[自动化与计算机技术—计算机科学与技术]