Research On Cloud Computing For Privacy Protection Algorithm Based On K-anonymity

With the rapid development of information era, the application of cloud computing in the study, life and work is more and more widely. The emergence of cloud computing saves a lot of cost and energy for the user, and now has been regarded as the key content of the new type of computer network system. However the development of cloud computing in security and in the application is out of sync, so far, there are still safe and secure threats need to face and solve in cloud computing security, especially in terms of privacy protection. Individual privacy is the most important information for people which is protected by law, but cloud computing is not perfect in privacy protection. Privacy information leaks emerge in endlessly, which all bring great threat to people’s life.Firstly, this paper introduces the relevant background knowledge of cloud computing, and implements researches and analyses on its basic definition, development process, characteristics, and some of the security threats in cloud computing environment, especially the security threat in cloud computing privacy protection. Through the comparison and analysis of several kinds of traditional privacy protection scheme, we find that the user’s privacy information is still possible to leak by linking attacks such as means of attack as further improved data mining technology. There are more and more methods, to makes it easier,to get date from network. Although the existing methods can solve the problem of privacy protection to a certain extent, in order to reduce the impact of the links attack on users’ privacy, this paper emphatically studies and improves the k-anonymity algorithm. K-anonymity algorithm can meet the requirements for protection of privacy in data publishing process in cloud computing environment. As a result, it is significantly important to use k-anonymity algorithm in cloud computing privacy protection.The two methods to implement k-anonymity technology is the generalization and hidden in k-anonymity protection model. It can cause a certain amount of loss of data quality that in the data table in the process of anonymization of data table. In this case, the user’s privacy information is likely to be protected. So we can say that there is contradiction between the degree of privacy and the data quality in the anonymous table, and we can’t get the two of them. In order to make the anonymous tables have higher practical value, finding the intermediate nodes between the degree of privacy and the data quality is the key of k-anonymity algorithm. The most critical process is to optimize the values of k. If the values is to large, it’s unable to make up for the loss of data quality; if it is too small, it will make the k-anonymity protection model can’t provide enough privacy to protect the privacy of customers, and then the K- anonymity privacy protection model will become meaningless. So, K value optimization process is the key process to make the k- anonymity table available, it is vital for the practical application of k- anonymity privacy protection model.Focus of this paper is to k- anonymity algorithm to optimize the k value, so that the user privacy information is anonymized after an anonymous table can effectively guarantee user privacy, but also to protect the quality of the user’s data is not much loss.In order to get the appropriate k value to anonymize raw data table, the first thing is to find the range of values of k under all of the conditions, and how to find the range of values of k is the emphasis of this study. First of all, makes theoretical research and experimental analysis to the relation between the change of the k value and the degree of privacy protection of anonymity table, and puts forward a constraint inequality of k. Then, this paper researches and analyzes the relation between the different k values in the table and anonymous data quality change, and puts forward another constraint inequality of k. Based on the two constraint inequalities, the scope of k meeting the requirements is derived. Finally, according to the different requirements of different users, the algorithm works out and selects the optimal value of k to implement the anonymization, and obtains the anonymity table which users require.