Private machine learning classification based on fully homomorphic encryption

Xiaoqiang Sun, Peng Zhang, Joseph K. Liu, Jianping Yu, Weixin Xie

    Research output: Contribution to journalArticleResearchpeer-review

    Abstract

    Machine learning classification is an useful tool for trend prediction by analyzing big data. As supporting homomorphic operations over encrypted data without decryption, fully homomorphic encryption (FHE) contributes to machine learning classification without leaking user privacy, especially in the outsouring scenario. In this paper, we propose an improved FHE scheme based on HElib, which is a FHE library implemented based on Brakerski's FHE scheme. Our improvement focuses on two aspects. On the one hand, we first use the relinearization technique to reduce the ciphertext size, and then the modulus switching technique is used to reduce the modulus and decryption noise. On the other hand, we need no relinearization and modulus switching if there is additive homomorphic or no homomorphic operation in the multiplicative ciphertext's next homomorphic operation. Homomorphic comparison protocol, private hyperplane decision-based classification and private Naïve Bayes classification are implemented by additive homomorphic and multiplicative homomorphic firstly. In our homomorphic comparison protocol, the number of interactions is reduced from 3 to 1. We choose the proposed FHE scheme to implement private decision tree classification. Simulation results show that the efficiency of our FHE scheme and implementation of private decision tree classification are more efficient than other two schemes.

    Original languageEnglish
    Pages (from-to)1-13
    Number of pages13
    JournalIEEE Transactions on Emerging Topics in Computing
    DOIs
    Publication statusAccepted/In press - 2019

    Keywords

    • Additives
    • decision tree
    • Decision trees
    • Encryption
    • fully homomorphic encryption
    • hyperplane decision-based
    • machine learning classification
    • Naïve Bayes
    • Privacy
    • privacy preserving
    • Protocols
    • Switches

    Cite this

    @article{09f7755154014d02a16a963ab93e8020,
    title = "Private machine learning classification based on fully homomorphic encryption",
    abstract = "Machine learning classification is an useful tool for trend prediction by analyzing big data. As supporting homomorphic operations over encrypted data without decryption, fully homomorphic encryption (FHE) contributes to machine learning classification without leaking user privacy, especially in the outsouring scenario. In this paper, we propose an improved FHE scheme based on HElib, which is a FHE library implemented based on Brakerski's FHE scheme. Our improvement focuses on two aspects. On the one hand, we first use the relinearization technique to reduce the ciphertext size, and then the modulus switching technique is used to reduce the modulus and decryption noise. On the other hand, we need no relinearization and modulus switching if there is additive homomorphic or no homomorphic operation in the multiplicative ciphertext's next homomorphic operation. Homomorphic comparison protocol, private hyperplane decision-based classification and private Na{\"i}ve Bayes classification are implemented by additive homomorphic and multiplicative homomorphic firstly. In our homomorphic comparison protocol, the number of interactions is reduced from 3 to 1. We choose the proposed FHE scheme to implement private decision tree classification. Simulation results show that the efficiency of our FHE scheme and implementation of private decision tree classification are more efficient than other two schemes.",
    keywords = "Additives, decision tree, Decision trees, Encryption, fully homomorphic encryption, hyperplane decision-based, machine learning classification, Naïve Bayes, Privacy, privacy preserving, Protocols, Switches",
    author = "Xiaoqiang Sun and Peng Zhang and Liu, {Joseph K.} and Jianping Yu and Weixin Xie",
    year = "2019",
    doi = "10.1109/TETC.2018.2794611",
    language = "English",
    pages = "1--13",
    journal = "IEEE Transactions on Emerging Topics in Computing",
    issn = "2168-6750",
    publisher = "IEEE Computer Society",

    }

    Private machine learning classification based on fully homomorphic encryption. / Sun, Xiaoqiang; Zhang, Peng; Liu, Joseph K.; Yu, Jianping; Xie, Weixin.

    In: IEEE Transactions on Emerging Topics in Computing, 2019, p. 1-13.

    Research output: Contribution to journalArticleResearchpeer-review

    TY - JOUR

    T1 - Private machine learning classification based on fully homomorphic encryption

    AU - Sun, Xiaoqiang

    AU - Zhang, Peng

    AU - Liu, Joseph K.

    AU - Yu, Jianping

    AU - Xie, Weixin

    PY - 2019

    Y1 - 2019

    N2 - Machine learning classification is an useful tool for trend prediction by analyzing big data. As supporting homomorphic operations over encrypted data without decryption, fully homomorphic encryption (FHE) contributes to machine learning classification without leaking user privacy, especially in the outsouring scenario. In this paper, we propose an improved FHE scheme based on HElib, which is a FHE library implemented based on Brakerski's FHE scheme. Our improvement focuses on two aspects. On the one hand, we first use the relinearization technique to reduce the ciphertext size, and then the modulus switching technique is used to reduce the modulus and decryption noise. On the other hand, we need no relinearization and modulus switching if there is additive homomorphic or no homomorphic operation in the multiplicative ciphertext's next homomorphic operation. Homomorphic comparison protocol, private hyperplane decision-based classification and private Naïve Bayes classification are implemented by additive homomorphic and multiplicative homomorphic firstly. In our homomorphic comparison protocol, the number of interactions is reduced from 3 to 1. We choose the proposed FHE scheme to implement private decision tree classification. Simulation results show that the efficiency of our FHE scheme and implementation of private decision tree classification are more efficient than other two schemes.

    AB - Machine learning classification is an useful tool for trend prediction by analyzing big data. As supporting homomorphic operations over encrypted data without decryption, fully homomorphic encryption (FHE) contributes to machine learning classification without leaking user privacy, especially in the outsouring scenario. In this paper, we propose an improved FHE scheme based on HElib, which is a FHE library implemented based on Brakerski's FHE scheme. Our improvement focuses on two aspects. On the one hand, we first use the relinearization technique to reduce the ciphertext size, and then the modulus switching technique is used to reduce the modulus and decryption noise. On the other hand, we need no relinearization and modulus switching if there is additive homomorphic or no homomorphic operation in the multiplicative ciphertext's next homomorphic operation. Homomorphic comparison protocol, private hyperplane decision-based classification and private Naïve Bayes classification are implemented by additive homomorphic and multiplicative homomorphic firstly. In our homomorphic comparison protocol, the number of interactions is reduced from 3 to 1. We choose the proposed FHE scheme to implement private decision tree classification. Simulation results show that the efficiency of our FHE scheme and implementation of private decision tree classification are more efficient than other two schemes.

    KW - Additives

    KW - decision tree

    KW - Decision trees

    KW - Encryption

    KW - fully homomorphic encryption

    KW - hyperplane decision-based

    KW - machine learning classification

    KW - Naïve Bayes

    KW - Privacy

    KW - privacy preserving

    KW - Protocols

    KW - Switches

    UR - http://www.scopus.com/inward/record.url?scp=85041666841&partnerID=8YFLogxK

    U2 - 10.1109/TETC.2018.2794611

    DO - 10.1109/TETC.2018.2794611

    M3 - Article

    SP - 1

    EP - 13

    JO - IEEE Transactions on Emerging Topics in Computing

    JF - IEEE Transactions on Emerging Topics in Computing

    SN - 2168-6750

    ER -