摘要
针对异构网络多维数据挖掘执行效率低、管理难度大和挖掘精度低等问题,提出基于粒度计算的异构网络多维数据挖掘方法。基于异构网络数据之间的差异化特性,建立具有数据共享、存储及管理功能的异构网络多维数据模型。依据决策逻辑理论定义异构网络多维数据模型中的多维数据信息表,利用模式搜索对数比率来粒化处理信息表中的结构粒,构造体现多维数据特征的粒度计算模型。引入六元组界定粒度计算模型,实现不同种类粒度的自适应计算,输出异构网络多维数据挖掘结果。通过实验结果可知,在图书馆推荐系统的异构网络中,该方法的挖掘精度高于97%,所推荐图书与用户的关联度高于0.950,挖掘性能较强。
Aiming at the problems of low execution efficiency,high management difficulty and low mining accuracy of multidimensional data in heterogeneous networks,a method of multidimensional data mining in heterogeneous networks based on granular computing is proposed.The differences between heterogeneous data are utilized to establish a multidimensional data model of heterogeneous network with data sharing,storage and management functions.The multidimensional data information table in the heterogeneous network multidimensional data model is defined according to the theory of decision logic,and the structure grain in the information table is granulated by using the logarithmic ratio of pattern search,and a granular computing model reflecting the characteristics of multidimensional data is constructed.The six-tuple defined granularity computing model is introduced to realize adaptive computing of different kinds of granularity and output multidimensional data mining results of heterogeneous networks.The experimental results show that in the heterogeneous network of the library recommendation system,the mining accuracy of this method is higher than 97%,and the correlation between the recommended books and users is higher than 0.950,and the mining performance is high.
作者
贾超贤
JIA Chaoxian(Huainan Vocational Technical College,Huainan Anhui 232001,China)
出处
《重庆科技学院学报（自然科学版）》
CAS
2023年第2期87-91,共5页
Journal of Chongqing University of Science and Technology：Natural Sciences Edition
基金
2021年度高等学校省级质量工程项目“计算机应用技术专业结构优化调整与专业改造”(2021ZYYH038)
2021年度淮南职业技术学院自然科学研究重点项目“针对UGC数据进行数据挖掘的研究与实现”(HKJ21-3)。
关键词
粒度计算
异构网络
多维数据
决策逻辑理论
六元组
granular computing
heterogeneous network
multidimensional data
decision logic theory
six-tuple
