A Novel Data Purification Algorithm Based On Outlier Mining

被引：0

作者：

Dong, Jianfeng ^{[1
]}

Wang, Xiaofeng ^{[2
]}

Hu, Feng ^{[1
]}

Xiao, Liyan ^{[3
]}

机构：

[1] Wuhan Univ, Sch Informat Management, Wuhan 430072, Peoples R China

[2] Huazhong Univ Sci & Technol, Sch Comp Sci, Wuhan, Peoples R China

[3] Jishou Univ, Coll Foreign Language, Zhang Jiajie, Peoples R China

来源：

HIS 2009: 2009 NINTH INTERNATIONAL CONFERENCE ON HYBRID INTELLIGENT SYSTEMS, VOL 3, PROCEEDINGS | 2009年

关键词：

data mining; outlier; data purification; exception set;

D O I：

10.1109/HIS.2009.231

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

this paper presents a data purification algorithm based on outlier mining. In order to implement the purifying of training data, we define the central bias function of complex events and dissimilarity function of event set, and put forward an exception set growth algorithm based on bias priority. Experiment proves that the algorithm can solve non-deterministic polynomial hard and control the algorithm complexity within polynomial complexity.

引用

页码：95 / +

页数：2