Outsourced privacy-preserving C4.5 decision tree algorithm over horizontally and vertically partitioned dataset among multiple parties

Li, Ye; Jiang, Zoe L.; Yao, Lin; Wang, Xuan; Yiu, Siu-Ming; Huang, Zhengan

doi:10.1007/s10586-017-1019-9

Cited by 27 publications

(19 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using additive secret-sharing and with the help of a computing server, the cloud builds the tree from the encrypted data via tailored secure-counting and comparison protocols. They improve similar works [109,110], by relaxing the need for parties to be online during the learning.…”

Section: Secure Multiparty Computation Du and Zhanmentioning

confidence: 94%

“…For optimization reasons, several works modify the quality metric used by the learning algorithm [5,17,34,38,53,54,112,114,122,149,178,183] (see §6). Overall in the literature, we observe a wide range of combinations on tree-types, tasks, data, and algorithms, as illustrated CART [34] CART-like [5] C4.5-like [168] CRT [116] ExRT [119] ExRT [62] CART [15,17,71,85,96,110,152,154,158,181] C4.5 [41] CART [172,192] CART-like [156] CRT [68,171] ID3/C4.5 [112] C45 [7,8,27] CART [9] C4.5 [16,26,79,183] CART [95,108,177] CRT [84] ID3/CRT [20] CRF …”

Section: Learning Algorithmmentioning

confidence: 99%

“…We identify five categories of PETs: (a) input randomization ( §6.1), (b) differential privacy-based solutions ( §6.2), (c) cryptographic approaches ( §6.3), (d) hardware-based solutions ( §6.4), and (e) hybrid solutions that combine the above ( §6.5). [41,55,56,76,96,160] FD (2 parties) [111,121,181] FD (2 parties) [80] Leader [32,37,153,166,186] Leader [72,104,156,187] Leader [190] Sequential [110,162] Aggregator [52,117,119,152,154,172] Aggregator…”

Section: Protection Mechanismmentioning

confidence: 99%

“…However, the special-role parties need to be available and to follow the protocol. In the offloading scenario, some works introduce an additional entity [107,110,123]. This helps with the computation of non-linear functions and comparisons through multi-party HE protocols among two non-colluding parties [25].…”

Section: Secure Multiparty Computation Du and Zhanmentioning

confidence: 99%

“…Remark. Note that in several works, the privacy is guaranteed by the presence of implicitly trusted thirdparties involved in the generation of cryptographic keys [107,110] and random shares used for the computations (e.g., Beaver triplets for multiplication operations in SMC) [179]. Similarly, hardware-based solutions presume the chip manufacturer is trusted and the attacker does not have access or control over the enclave [101,102].…”

Section: Threat Modelmentioning

confidence: 99%

See 4 more Smart Citations

SoK: Privacy-Preserving Collaborative Tree-based Model Learning

Chatel

Pyrgelis

Troncoso-Pastoriza

et al. 2021

Proceedings on Privacy Enhancing Technologies

View full text Add to dashboard Cite

Tree-based models are among the most efficient machine learning techniques for data mining nowadays due to their accuracy, interpretability, and simplicity. The recent orthogonal needs for more data and privacy protection call for collaborative privacy-preserving solutions. In this work, we survey the literature on distributed and privacy-preserving training of tree-based models and we systematize its knowledge based on four axes: the learning algorithm, the collaborative model, the protection mechanism, and the threat model. We use this to identify the strengths and limitations of these works and provide for the first time a framework analyzing the information leakage occurring in distributed tree-based model learning.

show abstract

Section: Secure Multiparty Computation Du and Zhanmentioning

confidence: 94%