Privacy Preserving Big Data Publication On Cloud Using Mondrian Anonymization Techniques and Deep Neural Networks

Layers Based Optimal Privacy Preservation of the On-premise Data Supported by the Dual Authentication and Lightweight on Fly Encryption in Cloud Ecosystem

“…9 The computational cost in terms of time complexity for re-encryption with block size 16, 64, 256 and 1024 for file chunking into data node in the cloud with rkey = 256 bit Figure 9 exhibits how the proposed system model behaves when it deals with reencryption for file chunking into the data node in the cloud with rKey of 256 bit size. The interpretation of the outcome of running time (ms) shows that cost of computation for file-chunking is quite lesser in the case of chunk size (16) for the formulated ACM-PPC re-encryption approach on cloud ecosystem, but when the chunk sizes increase from 64 to 1024, then the system yields marginally higher outcome of cost of computation. On average, the file-chunking of the re-encryption model for a block size of 16 provides costefficient performance with an average of 68% over the file chucking by the block size of 64, 256, and 1024.…”

“…The limitation of the data anonymizations on the Bigdata limits its use to balance data privacy and data usability. The authors [16] propose a variance of the anonymization as Mondrian-K anonymity and Deep Neural Network balance a tradeoff between the data utility and security as privacy preservation.…”

“…To an extent, the strong authentication system fails the attacker's access to the data, yet the data is vulnerable to the insider attacker, which demands an effective privacy preservation method to gain the confidentiality of the private data. The isolated studies discussed in [11][12][13][14][15][16] towards privacy preservation need to be complemented by integrating it to the strong authentication, and it is being observed that there are fewer studies towards including the encryption mechanism which balances a tradeoff between authentication process time minimization and security levels for preserving the confidentiality in the cloud ecosystem. Thereby it is realized the development of a collaborative and effective multi-layer authentication and privacy data preservation is highly desirable, where the light-weighted fly encryption for the distributed file systems structure in the cloud data nodes becomes feasible.…”

Towards Human-Aware D2D Communication

“…Despite not being the focus of this work, user-data privacy is a must to support applications and innovation, without jeopardizing individual rights and security [11]. We argue that using appropriate techniques to guarantee user privacy according to the data type is necessary.…”

Tactful Networking: Humans in the Communication Loop

“…In [169], authors discuss the privacy problem in big data, methods to protect data publishing, evaluate big data (e.g., velocity, volume, variety, variability) from a privacy perspective, and recommendations for future research. In [170], the authors find that existing anonymization techniques fail in the trade-off between data utility and privacy. They propose a Mondrian based k-anonymity approach combined with a Deep Neural Network-based framework.…”