Privacy-Preserving Periodic Frequent Pattern Model in AIoT Applications

Abstract

We begin with a sanitization strategy for concealing sensitive periodic frequent patterns in this study. The developed method employs the Term Frequency and Inverse Document Frequency (TF-IDF) to determine which transactions and objects should be sanitized based on user-defined sensitive periodic frequent patterns. Using the designed approach, it is possible to correctly and properly choose victim items in the transactional database for data sanitization.