ABSTRACT
In recent years, the data mining proficiencies have met a dangerous challenge due to the altered regarding
and concerns of the privacy that is, defending the secrecy of the vital and sore data. Different proficiencies
and algorithms have been already demonstrated for Privacy Preserving data mining, which could be
assorted in three common approaches: Data modification approach, Data sanitization approach and Secure
Multi-party Calculation approach. This paper demonstrates a Data modification– based Framework for
categorization and valuation of the privacy maintaining data mining techniques. Based on our model the
proficiencies are divided into two major groups, namely perturbation approach and anonymization
approach. Also in proposed model, eight functional criteria will be used to examine and analogically
judgment of the proficiencies in these two major groups. The suggested framework furnishes a good basis
for more accurate comparison of the given proficiencies to privacy maintaining data mining. In addition,
this framework permits distinguishing the overlapping quantity for different approaches and describing
modern approaches in this field.
Keywords: - Privacy Preserving Data Mining, Data Modification, Perturbation, Anonymization