摘要
目前大多数数据挖掘方法是从单关系中发现模式,而多关系数据挖掘(MRDM)则可直接从关系数据库的多表中抽取有效模式。MRDM可以解决原有命题数据挖掘方法不能解决的问题,它不仅有更强的信息表示能力,可以表示和发现更复杂的模式,还可以在挖掘进程中有效地利用背景知识来提高挖掘效率和准确率。近年来,借鉴归纳逻辑程序设计(ILP)技术,已经形成许多多关系数据挖掘方法,如关系关联规则挖掘方法、关系分类聚类方法等。
Most existing data mining approaches look for patterns in a single table, Multi-Relational Data Mining(MRDM) approaches directly look for patterns that involve multiple tables (relations) from a relational databases. MRDM approaches can solve some problems which is difficult to deal with for propositional data mining approaches. Apart from better expressiveness, MRDM approaches could also make full use of background knowledge in the process of discovery. In recently years, using ILP technology for reference, many popular data mining methods have been extended to the multi-relational case, such as relational association rules, relational classification approaches, etc.
出处
《计算机应用研究》
CSCD
北大核心
2006年第9期8-12,共5页
Application Research of Computers
基金
国家科技成果重点推广计划项目(2003EC000001)
关键词
多关系数据挖掘(关系数据挖掘)
归纳逻辑程序设计
关系分类回归
关系关联规则
基于距离的关系方法
Multi-Relational Data Mining(MRDM) ( Relational Data Mining, RDM)
Inductive Logic Programming (ILP)
Relational Classification and Regression
Relational Association Rules
Relational Distance-based Methods