|
基于粒度计算的异构网络多维数据挖掘方法 |
A Multidimensional Data Mining Method for Heterogeneous Networks Based on Granular Computing |
投稿时间:2022-07-20 修订日期:2022-07-20 |
DOI: |
中文关键词: 粒度计算 异构网络 多维数据 挖掘方法 决策逻辑理论 六元组 |
英文关键词: Granular computing Heterogeneous network Multidimensional data Mining method Decision logic theory Six-tuple |
基金项目:2021年淮南职业技术学院自然科学研究项目重点项目 《针对UGC数据进行数据挖掘的研究与实现》(编号:HKJ21-3);2021年高等学校省级质量工程项目省级《计算机应用技术专业结构优化调整与专业改造》(编号:2021zyyh038) |
|
摘要点击次数: 193 |
全文下载次数: 0 |
中文摘要: |
针对异构网络多维数据密度高、关联复杂导致挖掘精度低的问题,研究基于粒度计算的异构网络多维数据挖掘方法。利用异构数据间存在的差异化特性以及大数据网络存储设备,建立具有数据共享、存储以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粒化处理信息表内的结构粒,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。实验结果表明,将该方法应用于图书馆推荐系统的异构网络中,挖掘多维数据的精度高于97%,所推荐图书与用户的关联度高于0.95,挖掘性能高。 |
英文摘要: |
Aiming at the problem of low mining accuracy caused by high density and complex association of multidimensional data in heterogeneous networks, a multidimensional data mining method based on granular computing is studied. Using the differences between heterogeneous data and big data network storage devices, a heterogeneous network multidimensional data model with data sharing, storage and management functions is established. According to the theory of decision logic, the multidimensional data information table in the multidimensional data model of heterogeneous networks is defined, and the structural particles in the information table are granulated by using the logarithmic ratio of pattern search to construct a granular computing model that reflects the characteristics of multidimensional data. The six tuple defined granularity computing model is introduced to realize the adaptive computing of different kinds of granularity, and the results of multidimensional data mining in heterogeneous networks are output. The experimental results show that when this method is applied to the heterogeneous network of Library recommendation system, the accuracy of mining multidimensional data is higher than 97%, and the correlation between the recommended books and users is higher than 0.95, so the mining performance is high. |
View Fulltext
查看/发表评论 下载PDF阅读器 |
关闭 |