Improving accuracy of classification models induced from anonymized datasets
Information Sciences(2014)
摘要
•We present a new anonymization algorithm for privacy-preserving data publishing.•The non-homogeneous generalization is coupled with sensitive value distributions.•The predictive utility of the algorithm is measured on eight anonymized datasets.•The algorithm outperforms other methods with four classification techniques.
更多查看译文
关键词
Privacy preserving data publishing,Privacy preserving data mining,k-Anonymity,ℓ-Diversity,Non-homogeneous anonymization,Classification
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要