Optimization techniques for preserving privacy in data mining

• Published in: 2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT) pp. 1 - 6
• Main Authors: Devi, K.Renuka, Balasamy, K., Prathyusha, M., Jeevitha, R., Balasubramanie, P., Eswaran, Malathi
• Format: Conference Proceeding
• Language: English
• Published: IEEE 22.02.2023
• Subjects: Costs; Data mining; Data privacy; Genetic algorithm; Medical services; Metaheuristics; Optimization methods; Particle swarm optimization; Perturbation methods; Privacy; Reinforcement learning
• DOI: 10.1109/ICECCT56650.2023.10179655

Data mining is one of the significant area where it plays a predominant role in extracting important factors and trends from large volume of data. This covers various areas such as healthcare, education, entertainment, finance, e-commerce applications etc., The data mining domain has used a variety of algorithms, including supervised, unsupervised, semi-supervised, and reinforcement learning techniques. Under healthcare arena, it deals with huge amount of sensitive data such as patients' data such as their name, age, health records. Those sensitive data have been utilized by the intruders for extracting the original data and also became a prey for the authorized access. Hence, the privacy is one of the serious concern that should be addressed. Various privacy preserving in data mining (PPDM) techniques such as anonymization, perturbation, condensation and cryptographic methods are available to protect those data. In this paper, the optimization techniques such as Genetic algorithm(GA) under evolutionary method and Particle swarm optimization(PSO) under meta heuristic method have been discussed and how it plays an important part in providing more optimal results by securing those sensitive and important information from the unauthorized access.