期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2015
卷号：4
期号：2
页码：528-532
出版社：Shri Pannalal Research Institute of Technolgy
摘要：Data mining is a rapidly growing beneficial technology for getting useful information stored in huge data collection. Behind this data mining technique discrimination is one of the most important concepts when considering legal and ethical factors of privacy preservation. The information society contains services which can automatically and routinely collect very large data. This large data is used in various applications like insurance premium computation ,staff selection ,loan granting, education etc. For the matter of sensitivity caused due to the sensitive attribute, our system will introduce new concept of anti-discrimination techniques which contains the two tasks discrimination discovery and prevention for data mining .There are two types of discrimination one is direct discrimination and another is indirect discrimination. In the direct discrimination method, the decisions are made based on the sensitive attributes like age, race, marital status, disability, religion, gender. And in indirect discrimination method the decisions are made on the basis of non-sensitive attributes, which are strongly correlated with sensitive attributes. In this paper, we are presenting various discrimination detection and prevention methods along with direct and indirect discrimination. We will show how to remove the discrimination for any data sets which will be helpful to preserve the quality of data.
关键词：anti-discrimination ;data mining; direct and ; indirect discrimination prevention; privacy preservation; rule ; protection; rule generalization
