Application of Homomorphic Encryption in Machine Learning

Behera, Sagarika; Prathuri, Jhansi Rani

doi:10.1109/phdedits51180.2020.9315305

Cited by 14 publications

(8 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A hash function used by the Distributed Hash Table (DHT) to assign file ownership to network nodes which generates a key of 256 bits, which is enough to withstand the level 2 attack on the DRE. This system's encryption function hides the distance between two points in a database table; therefore, it must be determined which of the two points is closest to a query point q , and it must also be implemented [ 4 , 26 ]:

where ||p|| represents the Euclidean norm of p , represents the gradient system, and ||p||2 can be represented by pp . As a result, the problem of inequality can be broken down into a slew of gradient calculations.…”

Section: Methodsmentioning

confidence: 99%

Recommendation System for Privacy-Preserving Education Technologies

Yin

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Considering the priority for personalized and fully customized learning systems, the innovative computational intelligent systems for personalized educational technologies are the timeliest research area. Since the machine learning models reflect the data over which they were trained, data that have privacy and other sensitivities associated with the education abilities of learners, which can be vulnerable. This work proposes a recommendation system for privacy-preserving education technologies that uses machine learning and differential privacy to overcome this issue. Specifically, each student is automatically classified on their skills in a category using a directed acyclic graph method. In the next step, the model uses differential privacy which is the technology that enables a facility for the purpose of obtaining useful information from databases containing individuals’ personal information without divulging sensitive identification about each individual. In addition, an intelligent recommendation mechanism based on collaborative filtering offers personalized real-time data for the users’ privacy.

show abstract

Section: Methodsmentioning

confidence: 99%

Recommendation System for Privacy-Preserving Education Technologies

Yin

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…As the popularity of ML increases in the cloud computing environment, privacy and security of the data become a great concern for cloud users. To overcome these problems, the researchers have given different methods for the application of homomorphic encryption in machine learning in the paper [18][19][20][21][22][23].…”

Section: Relevant Workmentioning

confidence: 99%

FPGA-Based Acceleration of K-Nearest Neighbor Algorithm on Fully Homomorphic Encrypted Data

Behera,

Prathuri

2024

Cryptography

Self Cite

View full text Add to dashboard Cite

The suggested solution in this work makes use of the parallel processing capability of FPGA to enhance the efficiency of the K-Nearest Neighbor (KNN) algorithm on encrypted data. The suggested technique was assessed utilizing the breast cancer datasets and the findings indicate that the FPGA-based acceleration method provides significant performance improvements over software implementation. The Cheon–Kim–Kim–Song (CKKS) homomorphic encryption scheme is used for the computation of ciphertext. After extensive simulation in Python and implementation in FPGA, it was found that the proposed architecture brings down the computational time of KNN on ciphertext to a realistic value in the order of the KNN classification algorithm over plaintext. For the FPGA implementation, we used the Intel Agilex7 FPGA (AGFB014R24B2E2V) development board and validated the speed of computation, latency, throughput, and logic utilization. It was observed that the KNN on encrypted data has a computational time of 41.72 ms which is 80 times slower than the KNN on plaintext whose computational time is of 0.518 ms. The main computation time for CKKS FHE schemes is 41.72 ms. With our architecture, we were able to reduce the calculation time of the CKKS-based KNN to 0.85 ms by using 32 parallel encryption hardware and reaching 300 MHz speed.

show abstract

“…Behera et al [15] have implemented the Paillier cryptosystem which follows additive HE property. They started with a plain text which is encrypted using HE method and this encrypted data is then fed as input to a linear regression model.…”

Section: B Literature Surveymentioning

confidence: 99%

Effectiveness and Limitations of existing techniques for privacy preservation data mining (PPDM) for Medical Data

Kulkarni,

Sujatha

2023

AJCT

View full text Add to dashboard Cite

Background & Objective: This survey paper examines the application of Federated Learning (FL) and secure Multiparty Computing (MPC) considering medical data privacy, further provides an overview of FL and MPC techniques and discusses their strengths and weaknesses. This also covers other techniques like Homomorphic Encryption, data masking, differential privacy for its efficiency, and limitations. Methods: Eligibility Criteria: For this survey PRISMA[1] framework was employed and popular electronic databases like IEEE, Computers and Security, Bioinformatics, and Google Scholar were scanned along with government websites and research web pages. Papers published between 2018 and 2023 and written in English are considered for the survey. This survey further provides direction towards the future research and potential challenges in the deployment of these techniques at a scale. Results: The search was restricted for the period between 2018 and 2023. Initially, ~100 papers were shortlisted, and after a though review of each paper, finally selected ~35 for this work. However, papers from earlier years are also included as they are found to be relevant for this study. Fig4 below describes the article selection process. Apart from the research papers, government websites are also referred for information on various laws, regulations, and compliance for the privacy of the patient's information. Acts like HIPPA[7], GDPR[8], PDP[9] are reviewed thoroughly to ensure all the aspects are covered to present this survey. Conclusion: The information used in the data mining process comes from data, and such data include personal data about people. Since knowledge is derived from data, the main goal of privacy-preserving data mining is the development of algorithms to conceal or protect certain sensitive information so that it cannot be disclosed to unintended users, hackers, or intruders. Various techniques like homomorphic encryption, differential privacy, data masking, data aggregation, multiparty computation is studied, one single technique may not be sufficient to provide an end-to-end solution. In the future work, we would like to explore on multiple techniques to build a secured ML model.

show abstract

Application of Homomorphic Encryption in Machine Learning

Cited by 14 publications

References 2 publications

Recommendation System for Privacy-Preserving Education Technologies

Recommendation System for Privacy-Preserving Education Technologies

FPGA-Based Acceleration of K-Nearest Neighbor Algorithm on Fully Homomorphic Encrypted Data

Effectiveness and Limitations of existing techniques for privacy preservation data mining (PPDM) for Medical Data

Contact Info

Product

Resources

About