Inference Mining using Direct and Indirect Discrimination Prevention in Data Mining

Data Mining is an essential and flourishing technology to extract the relevant and useful information hidden in the large collections of data. Privacy preservation in data mining is an important issue when considering the legal and ethical aspects of data mining. Discrimination is one of the facts that pave the way for negative perceptions in the data mining. Direct and Indirect discrimination are the two types of discrimination which involves in sensitive and non sensitive attributes. In this paper, our effort is to prevent the discrimination from both types and proposing the inference mining technique which allows the decision making process whether to accept or deny the request. Keywords— DD-Direct Discrimination; IDD-Indirect Discrimination; PPDM-Privacy Preserving Data Mining; IM-Inference Mining; PD-Potential Discrimination; PND-Potential Non Discrimination