A SHORT EVALUATION ON DIFFERENTIAL PRIVACY FOR PRIVACY PRESERVATION OF BIG DATA WITH FUTURE RESEARCH PERSPECTIVES
Owing to many communication technology devices, processes, and electronic information, such as the cloud, corporate data, social networks, internet activities, personal archives, and sensors, the information age has exponentially grown these types of data. Properly managing this enormous and diverse data is the most difficult problem. One of these significant and varied sorts of data is big data. The major big data problems are privacy and security because of its huge size. There is a risk because privacy could be revealed at any level in managing big data. The characteristics of unsuited methods of big data, such as anonymization and encryption, have been established to prevent the privacy of enormous datasets. Thus, this research discusses various literature works on differential privacy preservation for big data. It also reveals the intelligent protocols' methodologies, including machine learning and deep learning. Moreover, the survey part is also analyzed by exploring the chronological review of privacy preservation for big data, utilized big data, performance metrics for evaluation, and implementation platforms. At last, the challenges observed in traditional research works for privacy preservation using differential privacy for big data are discussed, and the future perspectives on promoting the utilization of big data are also explained.
Privacy Preservation of Big Data; Differential Privacy; Methodologies of Big Data; Performance Metrics; Implementation Tools; Dataset Details; Algorithmic Classification