Explore chapters and articles related to this topic
FGUGChain
Published in E. Golden Julie, J. Jesu Vedha Nayahi, Noor Zaman Jhanjhi, Blockchain Technology, 2020
A. Anasuya Threse Innocent, G. Prakash
When the number of parties involved are three or more, to compute a common functionality without revealing their private data, it is termed as Secure Multiparty Computation, or simply as Multiparty Computation, denoted by MPC. Election can be considered as an example of multiparty computation; voters want that their votes (and their identity) be kept secret, and at the same time it has to be counted. Another example is conducting a study of a new disease without revealing the individual test reports of patients or their identities. The baseline is, computation has to be done on private data without exposing them, and the outcome should reveal only the desired output, and whatever minimal information leaked by it. A number of day-to-day life problems ranging from as simple as coin tossing and mutual agreement to complex applications such as e-voting, electronic auctions, private data retrieval, analysis of sensitive information to conduct research on it without exposing them, privacy preserving biometric identification, private editing in cloud, etc. can be solved by secure computation.
Quality medical data management within an open AI architecture – cancer patients case
Published in Connection Science, 2023
Mirjana Ivanovic, Serge Autexier, Miltiadis Kokkonidis, Johannes Rust
In line with data preparation for application of advanced AI/ML approaches is privacy preserving as very prominent issue. Apart from a range of already existing methods necessity to leverage privacy-enhancing technologies is essential to reap the benefits of AI/ML while minimising the risk of data violations (Link 5). Apart from traditional methods, emerging privacy-enhancing technologies that will significantly increase privacy and security of sensitive medical and health data in the future include: Differential privacy, Homomorphic encryption, Secure multi-party computation, Zero-knowledge proofs, and so on. Within ASCAPE architecture, we verified effectiveness of Differential privacy and Homomorphic encryption approaches as guarantee for privacy preserving of sensitive patients’ data.
Deep Learning: Differential Privacy Preservation in the Era of Big Data
Published in Journal of Computer Information Systems, 2023
The human action recognition model produces an enormous dataset from IoT for facilitating data sharing among different data providers. With this regard, Kwabena Owusu-Agyemeng et al.58 proposed multi-scheme differential privacy (MSDP). That proposed scheme was developed based on Secure Multi-party Computation (SMC) and -differential privacy. The CNN structure was modified with the max pooling and rectified linear units (ReLU) to improve classification accuracy. Moreover, the polynomial approximation of batch normalization was integrated with ReLU when substituting max pooling with a sum pooling. These methods show the benefits of DL in big data analysis and also the drawbacks of that methods. This method verified that the DL methods had provided considerable performance efficacy in a big data environment rather than K-anonymity and cryptographic methods. The methods mentioned above and their advantages are illustrated in Table 4. Several approaches for big data privacy are depicted in Figure 10.
BFS: A blockchain-based financing scheme for logistics company in supply chain finance
Published in Connection Science, 2022
Jia Fu, Bangcan Cao, Xiaoliang Wang, Pengjie Zeng, Wei Liang, Yuzhen Liu
In terms of privacy protection in blockchain, the commonly used privacy protection algorithms in blockchain technology are secure multi-party computation platform, Homomorphic Encryption, etc., and these privacy protection algorithms are also widely used (Liang et al., 2021). Wang, She, et al. (2021) provided an effective solution for blockchain technology with privacy by performing a new optimisation of the secure computation protocol. Liang, Zhang, et al. (2020) established a mathematical model based on homomorphic encryption using blockchain and smart contracts, followed by the design of algorithms including blockchain generation, homomorphic chain encryption/decryption and smart contracts. For supply chain finance scenario, Ma et al. (2019) gave a scenario where the writing of smart contracts in the Fabric platform makes it mandatory for organisations joining the blockchain to be authenticated under the Fabric-CA organisation, preventing illegal organisations from joining the network to steal users’ privacy and ensuring data privacy. Liang, Fan, et al. (2020) proposed a secure data storage and recovery scheme based on blockchain networks that can quickly recover data from failed nodes while protecting user data privacy.