Research On Privacy Preserving Data Publishing Technologies

Modern society has experienced three times of science and technology revolution, every time the outbreak of the revolution of science and technology bring productivity will be a great progress. The third revolution of science and technology, the network development gradually toward popularization, the information sharing and resource efficiency has been more and more attention. Protect people’s privacy issue has become a focus of the attention, also is an important research direction. In the process of publishing data, if only to explicitly identifier can accurately determine the identity of the users to encrypt or delete, data privacy protection effect is not good, the attacker can still through the use of published data set identifier attributes and foreign data collection for joint exercise, the use of multiple data sets the link operation, so as to determine the individuals want to be to protect the privacy of data information. In the release of the data privacy protection, a privacy protection method is the most important is the K-anonymous technology. But how in the multiple attribute set to obtain the optimal K-anonymous is a NP hard problem, and so on of K-anonymous research focused on: First, published data collection anonymous operation control in a reasonable time complexity; Second, at the same time can get higher anonymity degree.This paper analyses the current K-anonymous algorithm, and summarizes the advantages and disadvantages of these methods are. Grab the paper focuses on use of my data set to validate K-anonymous algorithm, and put forward a new kind of relative equilibrium entropy-Incognito algorithm. To solve these problems, the thesis do the main work is as follows:1. Proposed E-Incognito algorithm. For related research status quo of data privacy protection, related knowledge of privacy protection were summarized. Detailed introduces the concept of the K-anonymous technology characteristic, development and application, possible attacks, the research status and research hot spot, and the common techniques of K-anonymous done in-depth research, the new proposed entropy and classic algorithms Incognito algorithm is an algorithm combining the entiopy-Incognito algorithm. The algorithm in the loss of time and achieve a good balance between accuracy of anonymous.2. Access to data sets. Data set to obtain concrete method is to use the programming language python web crawler. Web crawler through own module, the analysis of tree structure of a web page, from the given URL crawl web data that we need. The paper obtain a total of11426data.These data throughout the paper.3. The experimental validation. Using the data set to K-anonymous attribute generalization algorithm based on entropy entropy-Incognito, classic algorithm Incognito and time consumption and anonymous precision experiments, and compare the result of the experiment, the evaluation algorithm is introduced. Finally, the analysis and comparison of experimental results show that the rationality and validity of the proposed method in this paper, the test data of the quality and performance.