首页 | 本学科首页   官方微博 | 高级检索  
     检索      

匿名数据的KCNN-SVM分类技术
引用本文:张晓琳,张日初,刘立新.匿名数据的KCNN-SVM分类技术[J].郑州大学学报(理学版),2011,43(2).
作者姓名:张晓琳  张日初  刘立新
作者单位:内蒙古科技大学,信息工程学院,内蒙古,包头,014010
基金项目:国家社会科学基金资助项目,编号07XTQ003; 内蒙古自然科学基金重点资助项目,编号20080404Zd21; 教育部“春晖计划”基金资助项目,编号Z2009-1-01024
摘    要:提出一种基于匿名数据分类法的KCNN-SVM分类法.把匿名数据建模成不确定数据,实现用匿名数据建立分类模型.同时发布匿名数据的QI属性统计信息,更好地操纵匿名数据.KCNN-SVM分类法改进了SVM分类对于混淆点的泛化能力.结果表明,在实现了匿名数据分类的同时,分类精度也得到了提高.

关 键 词:隐私保护  匿名数据分类  KCNN-SVM

KCNN-SVM Classification for Anonymized Data
ZHANG Xiao-lin , ZHANG Ri-chu , LIU Li-xin.KCNN-SVM Classification for Anonymized Data[J].Journal of Zhengzhou University:Natural Science Edition,2011,43(2).
Authors:ZHANG Xiao-lin  ZHANG Ri-chu  LIU Li-xin
Institution:ZHANG Xiao-lin,ZHANG Ri-chu,LIU Li-xin(School of Information Engineering,Inner Mongolia University of Science and Technology,Baotou 014010,China)
Abstract:A classification approach called KCNN-SVM based on anonymized data classification was presented.This method created anonymized data as uncertain data and realized creating classification model using anonymized data.Meanwhile,QI-statistics information of anonymized data to manipulate anonymized data even better was released.The generalization ability of confusing point of SVM was improved.Anonymized data classification was realized and the classification accuracy was improved.
Keywords:privacy preservation  classification for anonymized data  KCNN-SVM  
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号