Clustering-Anonymity Method for Privacy Preserving Table Data Sharing

In the era of big data, the open sharing of government data has increasingly attracted the attention of governments. However, there is privacy leakage risk in the government’s data sharing. For the scene of sharing the table data, this paper proposes a approach for privacy-preserving data sharing in this paper based on anonymity clustering. Firstly, we preprocess the data table, and the records in the table are clustered by k-mediods clustering algorithm. The data table is divided into multiple sub-tables according to the distance between records. Then, the data records in the sub-table are divided based on the information loss parameter value, and the anonymous table data is adjusted so that the sensitive attribute values in the equivalence class are different. Last, Laplace noises are added to the value of sensitive attribute to ensure the privacy of the shared data. Compared with the classical k-anonymous MDAV algorithm in execute time, information loss and information entropy, the experimental results show that the proposed algorithm can reduce the operating time, improve the privacy protection to some extent, and has certain availability from the three aspects.