A review on Privacy Preservation and Collaborative Data Mining

Amit Kumar

doi:10.24297/ijct.v14i12.1777

Abstract

Privacy preservation is major issue in current data transmission over internet and cloud network. For the integrity and security of data various methods are used such as cryptography, data transformation, Steganography, watermarking and many more method. In consequence of all these method some data mining technique is used. The data mining technique provide Varity of algorithm for privacy preservation. The collaborative data mining technique used different agent method for the integrity of security of data during transmission. Issues about privacy-preserving data mining have emerged globally, but still the main problem is that non- sensitive information or unclassified data, one is able to infer sensitive information that is not supposed to be disclosed. Data collection is a necessary step in data mining process. Due to privacy reasons, collecting data from different parries becomes difficult. In this paper presents the review of privacy persevering technique used data mining.

Highlights

In current decade the integrity and security of data over internet is challenging task
The rule mining technique is strong approach of data mining used for privacy preservation[1,3]
Data mining a nontrivial extraction of novel, implicit, and actionable knowledge from large data sets is an evolving technology which is a direct result of the increasing use of computer databases in order to store and retrieve information effectively .It is known as Knowledge Discovery in Databases (KDD) and enables data exploration, data analysis, and data visualization of huge databases at a high level of abstraction, without a specific hypothesis in mind

Summary

Introduction

In current decade the integrity and security of data over internet is challenging task. In order to preserve privacy, passenger information records can be de-identified before the records are shared with anyone who is not permitted directly to access the relevant data[4,5] This can be done by removing from the dataset unique identity fields, such as name and passport number. Even though if this information is deleted, there are still other forms of information both personal and behavioural (e.g. date of birth, zip code, gender, number of children, number of calls, number of accounts) that, when connected with other available datasets, could recognize subjects To avoid these types of violations, we require various data mining algorithms for privacy preserving[7].

Objectives

Methods

Conclusion