Research On Privacy Protection Method For Dynamic Data

With the rapid development of Internet technology,data mining is widely used in many fields such as finance,healthcare,and retail.By analyzing the massive data and extracting the effective information and knowledge hidden in it,it can be effectively utilized.At the same time,privacy has become an urgent problem in the field of big data.Once a malicious user obtains private information that other users are not willing to let others know,they can use this information to carry out certain malicious acts,thereby causing economic losses or other aspects to other users.Therefore,how to protect users' private information while realizing efficient use of data has become a hot research direction in the field of big data.In the big data environment,privacy protection technology faces the following challenges:(1)The characteristics of data in the big data environment are constantly changing.Some traditional privacy protection technologies cannot be solved well,and data producers do not automatically participate in privacy.(2)The diversity of big data increases the risk of user privacy leakage during the process of multi-source data fusion.Due to the wide distribution of data,there may be some correlation in multiple data sets.In this case,the risk of privacy leakage after multi-data fusion is greater than the risk of privacy leakage in a single data set.In view of the problems in the above big data environment,this paper conducts a certain research on the privacy protection methods of dynamic data in the big data environment.The main research work of this paper is as follows:(1)Aiming at the problem of privacy leakage in the process of user location service,this paper proposes a location privacy protection method based on query fragmentation with user collaboration.The query is initiated by using the anchor information to replace the real location information of the user,and the content in the query request process is fragmented and sent to the server by other users.After the location server collects all the requests,it reorganizes and returns the query result obtained by the incremental neighbor algorithm according to the user anchor information to the user.This process reduces the risk of user disclosure of private information during changing location services.(2)For the problem of increasing the risk of user privacy information leakage in the process of multi-data fusion association analysis in big data environment,this paper constructs the ontology of mining association rules and discovers multipleontology through semantic-based ontology mapping.The method combines the output mapping table to integrate multiple ontology,and it analyzes the association rules implicit in the global ontology,and it uses the information increment method to calculate the correlation between attributes as the basis for discovering the implicit relationship.For the implicit relationship found in the multi-database environment,the privacy protection is implemented by using the fuzzy-based association rule hiding algorithm.The algorithm is based on a heuristic association rule hiding strategy,which reduces the side effects of sensitive association rules hiding on the original data stream by generalizing the data.