Hiding Fuzzy Sequential Patterns by a Multi-Objective Scheme

Improvements in technology have led to generating and mining huge amounts of data. Sequence mining is a research area of data mining that aims to extract useful knowledge from raw databases. Although transactions with numerical information are usually seen in real-world applications like bank records, customer baskets and network traffic records, and efforts for mining sequential patterns with quantitative values are reported in the technical literature, the issue of privacy of sensitive knowledge in the form of fuzzy sequential patterns has not received much attention. This paper addresses the problem of fuzzy sequential pattern hiding and proposes a multi-objective optimization procedure with few distortions which leads to the preserving of database fidelity. The proposed algorithm can also work with fuzzy databases and more details are handled by fuzzy sets. Distortions, fidelity and time-consumption of the proposed algorithm are three criteria used to compare the performance of it with the existing algorithms.