电子学报 ›› 2018, Vol. 46 ›› Issue (5): 1234-1239.DOI: 10.3969/j.issn.0372-2112.2018.05.032

• 科研通信 • 上一篇    下一篇

概念的属性约简及异构数据概念漂移探测

邓大勇1,3, 卢克文1, 黄厚宽2, 邓志轩1   

  1. 1. 浙江师范大学数理与信息工程学院, 浙江金华 321004;
    2. 北京交通大学计算机与信息技术学院, 北京 100044;
    3. 浙江师范大学行知学院, 浙江金华 321004
  • 收稿日期:2016-12-16 修回日期:2017-03-21 出版日期:2018-05-25
    • 作者简介:
    • 邓大勇 男,1968年出生,副教授,博士,现为浙江师范大学行知学院教师,主要研究方向为粗糙集、粒计算、数据挖掘等.Email:dayongd@163.com.
    • 基金资助:
    • 国家自然科学基金 (No.61473030); 浙江省自然科学基金 (No.LY15F020012); 浙江师范大学网络空间安全浙江省一流学科

Attribute Reduction for Concepts and Concept Drifting Detection in Heterogeneous Data

DENG Da-yong1,3, LU Ke-wen1, HUANG Hou-kuan2, DENG Zhi-xuan1   

  1. 1. College of Mathematics, Physics and Information Engineering, Zhejiang Normal University, Jinhua, Zhejiang 321004, China;
    2. School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China;
    3. Xingzhi College, Zhejiang Normal University, Jinhua, Zhejiang 321004, China
  • Received:2016-12-16 Revised:2017-03-21 Online:2018-05-25 Published:2018-05-25
    • Supported by:
    • National Natural Science Foundation of China (No.61473030); National Natural Science Foundation of Zhejiang Province,  China (No.LY15F020012); First-class Discipline of Zhejiang Province - Cyberspace Security of Zhejiang Normal University

摘要: 粗糙集是粒计算的一种重要方法,数据异构性是大数据的一种特征.针对异构数据问题,探索了粗糙集属性约简的本质,提出了概念属性约简的定义,它兼容值约简、Pawlak约简和并行约简.探究了概念属性约简的性质,提出了异构数据的属性约简方法和概念漂移探测方法.理论分析和示例表明了这些方法的有效性.为粗糙集、粒计算融入大数据的时代潮流提供了一种新方法.

关键词: 粒计算, F-粗糙集, 属性约简, 异构数据, 概念漂移

Abstract: Rough set theory is one of important methods of granular computing,and data heterogeneities are one of remarkable characteristics in big data.For data heterogeneities,we define attribute reduction for concepts after investigating intrinsic quality of attribute reducts,which can contain value reducts,Pawlak attribute reducts and parallel reducts.After investigating properties of concept-attribute-reduction,we present a new method to reduce redundant attributes and a new method to detect concept drift for heterogeneous concepts.Theoretical analysis and examples show that these methods are valid.This work provides a new type way for rough set theory and granular computing to integrate into big data.

Key words: granular computing, F-rough sets, attribute reduction, heterogeneous data, concept drifting

中图分类号: