Graph based Hiding of Sensitive Knowledge
Panteleimon Krasadakis, Giuseppe Futia, Vassilios S. Verykios, Evangelos Sakkopoulos · 2023
The Frequent Itemset Hiding Problem (FIHP) is an important research topic in the field of Privacy-Preserving Data Mining. Knowledge Graphs (KGs) have also recently emerged as a popular data representation model for various applications. However, the integration of KGs and FIHP is a relatively unexplored area of research. In this paper, we propose a novel approach to address the FIHP by incorporating Community Detection graph-based algorithms to hide sensitive knowledge in our data. Our approach preserves the structure and utility of the KG and efficiently produces the set of Itemsets that must be hidden. We evaluate our approach on popular FIHP datasets and demonstrate its effectiveness in terms of time efficiency.