KIDS:K-anonymization data stream base on sliding window
Junwei Zhang, Jing Yang, Jianpei Zhang, Yongbin Yuan · 2010
In a wide range of fields, data arrive in the form of high speed and huge data streams, and accompanying risks of disclosure of privacy. Most of previous studies about privacy preserving, such as k-anonymity methods, are excellent and effective, however, focus on static data sets. In this paper, we study a novel framework KIDS (K-anonymIzation Data Stream base on sliding window) to solve this problem by continuously k-anonymity on the sliding window. KIDS protects privacy of data stream well and considers the distribute density of data in data stream, thereby improve usefulness of data largely. Our theoretical analysis and experimental results show that we can receive more accurate data mining results by KIDS with high efficiency.