一种改进的基于概念格的数据挖掘算法  

An Improved Concept Lattice-Based Data Mining Algorithm

作　　者：李志坚[1,2] 莫建麟[1,2] 

机构地区：[1]电子科技大学,成都610054 [2]阿坝师范高等专科学校,四川汶川623002

出　　处：《重庆师范大学学报（自然科学版）》2013年第2期92-95,共4页Journal of Chongqing Normal University:Natural Science

基　　金：四川省自然科学基金(No.11ZB152)

摘　　要：为解决多维数据模型与关系数据模型之间的双向数据系统查询、数据清洗、数据转换、实现集中和分发数据的准确性与一致性等问题,通过对概念格的相关研究,将全局数据挖掘与局部数据挖掘相结合,提出一种改进的基于局部信息的全局概念格的数据挖掘算法,并将挖掘过程分解为ETL(Extraction-Transformation-Loading)动作,结合ETL处理工作流,实现并行分布式海量数据的时序挖掘。实验证明,该算法对增强数据加工能力具有一定的实用性。In order to solve the multidimensional data model and relational data model, query between the two-way data system, data cleansing, data conversion, distributed data accuracy and consistency control problem, this paper described the concept of grid-related, the global data mining combined with local data mining is proposed based on local information based on the concept of a global grid of data mining algorithm, and the mining process was divided into ETL action, combined with the ETL process workflow, using amounts of data distributed parallel sequence mining. Experiments show that the algorithm has a good effect on enhanced data processing capability.

关 键 词：数据 概念格 数据挖掘 集成 时序挖掘 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]