Discrimination like privacy is a big issue when legal and ethical aspects of Data mining are considered. Most people don’t like to be discriminated for their gender, religion, nationality, age and so on, especially when those attributes are needed for making decisions. Decisions like giving them a job, loan, insurance, etc. Hence it is highly desirable to discover such potential biases and eliminating them from the training data without harming their decision-making utility .Therefore antidiscrimination techniques including discrimination discovery and prevention have been introduced in data mining. Discrimination prevention consist inducing patterns which do not lead to discriminatory decisions even if the original training datasets are inherently biased. So By focusing on the discrimination prevention, we present a group of pre-processing discrimination prevention methods with different features of each approach and how These approaches deal with direct or indirect discrimination.