Personalized trajectory privacy data publishing scheme based on differential privacy

The proliferation of smart devices with location-based services has significantly facilitated people's lives and generated a large amount of trajectory data. Analyzing this data can contribute to societal development, such as the construction of public facilities and intelligent transportation systems. But illegal leakage of data poses a serious threat to individual privacy within the released data. Currently, differential privacy technology has emerged as a rigorous and standardized privacy protection framework widely applied in trajectory data publishing. However, existing methods often suffer from either excessive privacy protection or insufficient protection of individual privacy. Therefore, this paper proposes a personalized trajectory privacy data protection scheme based on differential privacy (DP_SR). The scheme combines TF-IDF statistics and designs personalized exponential noise to protect the sensitive personal data in each trajectory, achieving personalized privacy protection. Then an RTF-tree is constructed, and differential privacy techniques are employed to safeguard the security of the entire trajectory dataset. Experimental results on two real trajectory dataset demonstrate that the proposed scheme achieves a better balance between privacy protection and data utility compared with state-of-the-art algorithms.