{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,8,2]],"date-time":"2025-08-02T17:34:46Z","timestamp":1754156086006,"version":"3.41.2"},"reference-count":23,"publisher":"Emerald","issue":"2","license":[{"start":{"date-parts":[[2023,7,10]],"date-time":"2023-07-10T00:00:00Z","timestamp":1688947200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.emerald.com\/insight\/site-policies"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IJWIS"],"published-print":{"date-parts":[[2023,7,12]]},"abstract":"<jats:sec>\n<jats:title content-type=\"abstract-subheading\">Purpose<\/jats:title>\n<jats:p>Recently, deep learning (DL) has been widely applied in various aspects of human endeavors. However, studies have shown that DL models may also be a primary cause of data leakage, which raises new data privacy concerns. Membership inference attacks (MIAs) are prominent threats to user privacy from DL model training data, as attackers investigate whether specific data samples exist in the training data of a target model. Therefore, the aim of this study is to develop a method for defending against MIAs and protecting data privacy.<\/jats:p>\n<\/jats:sec>\n<jats:sec>\n<jats:title content-type=\"abstract-subheading\">Design\/methodology\/approach<\/jats:title>\n<jats:p>One possible solution is to propose an MIA defense method that involves adjusting the model\u2019s output by mapping the output to a distribution with equal probability density. This approach effectively preserves the accuracy of classification predictions while simultaneously preventing attackers from identifying the training data.<\/jats:p>\n<\/jats:sec>\n<jats:sec>\n<jats:title content-type=\"abstract-subheading\">Findings<\/jats:title>\n<jats:p>Experiments demonstrate that the proposed defense method is effective in reducing the classification accuracy of MIAs to below 50%. Because MIAs are viewed as a binary classification model, the proposed method effectively prevents privacy leakage and improves data privacy protection.<\/jats:p>\n<\/jats:sec>\n<jats:sec>\n<jats:title content-type=\"abstract-subheading\">Research limitations\/implications<\/jats:title>\n<jats:p>The method is only designed to defend against MIA in black-box classification models.<\/jats:p>\n<\/jats:sec>\n<jats:sec>\n<jats:title content-type=\"abstract-subheading\">Originality\/value<\/jats:title>\n<jats:p>The proposed MIA defense method is effective and has a low cost. Therefore, the method enables us to protect data privacy without incurring significant additional expenses.<\/jats:p>\n<\/jats:sec>","DOI":"10.1108\/ijwis-03-2023-0050","type":"journal-article","created":{"date-parts":[[2023,7,7]],"date-time":"2023-07-07T01:52:29Z","timestamp":1688694749000},"page":"61-79","source":"Crossref","is-referenced-by-count":3,"title":["Output regeneration defense against membership inference attacks for protecting data privacy"],"prefix":"10.1108","volume":"19","author":[{"given":"Yong","family":"Ding","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Peixiong","family":"Huang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hai","family":"Liang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Fang","family":"Yuan","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Huiyong","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"140","published-online":{"date-parts":[[2023,7,10]]},"reference":[{"key":"key2023071009261008200_ref001","first-page":"343","article-title":"GAN-Leaks: a taxonomy of membership inference attacks against generative models","year":"2020","journal-title":"Computer and Communications Security"},{"journal-title":"IEEE Transactions on Intelligent Transportation Systems","article-title":"CAMRL: a joint method of channel attention and multidimensional regression loss for 3D object detection in automated vehicles","year":"2022","key":"key2023071009261008200_ref002"},{"journal-title":"IEEE Transactions on Intelligent Transportation Systems","article-title":"PPO2: location privacy-oriented task offloading to edge computing using reinforcement learning for intelligent autonomous transport systems","year":"2022","key":"key2023071009261008200_ref003"},{"issue":"6","key":"key2023071009261008200_ref004","doi-asserted-by":"crossref","first-page":"1725","DOI":"10.1109\/TCSS.2022.3178416","article-title":"The joint method of triple attention and novel loss function for entity relation extraction in small data-driven computational social systems","volume":"9","year":"2022","journal-title":"IEEE Transactions on Computational Social Systems"},{"issue":"1","key":"key2023071009261008200_ref005","first-page":"1","article-title":"Research progress and challenges of membership inference attacks in machine learning. (in Chinese)","volume":"12","year":"2022","journal-title":"Operations Research and Blurring"},{"first-page":"770","article-title":"Deep residual learning for image recognition","year":"2016","key":"key2023071009261008200_ref006"},{"key":"key2023071009261008200_ref007","first-page":"1","article-title":"Membership inference attacks on machine learning: a survey","volume":"54","year":"2021","journal-title":"ACM Computing Surveys (CSUR)"},{"first-page":"2261","article-title":"Densely connected convolutional networks","year":"2017","key":"key2023071009261008200_ref008"},{"year":"2021","key":"key2023071009261008200_ref009","article-title":"Practical blind membership inference attack via differential comparisons"},{"first-page":"259","article-title":"MemGuard: defending against black-box membership inference attacks via adversarial examples","year":"2019","key":"key2023071009261008200_ref010"},{"issue":"1","key":"key2023071009261008200_ref011","doi-asserted-by":"crossref","first-page":"27988","DOI":"10.1038\/srep27988","article-title":"Multi-class texture analysis in colorectal cancer histology","volume":"6","year":"2016","journal-title":"Scientific Reports"},{"first-page":"7871","article-title":"BART: denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension","year":"2020","key":"key2023071009261008200_ref012"},{"first-page":"634","article-title":"Machine learning with membership privacy using adversarial regularization","year":"2018","key":"key2023071009261008200_ref013"},{"year":"2017","key":"key2023071009261008200_ref014","article-title":"Semi-supervised knowledge transfer for deep learning from private training data"},{"issue":"3","key":"key2023071009261008200_ref015","doi-asserted-by":"crossref","first-page":"446","DOI":"10.1108\/DTA-05-2020-0108","article-title":"Classification of electrocardiogram signal using an ensemble of deep learning models","volume":"55","year":"2021","journal-title":"Data Technologies and Applications"},{"year":"2019","key":"key2023071009261008200_ref016","article-title":"ML-leaks: model and data independent membership inference attacks and defenses on machine learning models"},{"issue":"5","key":"key2023071009261008200_ref017","doi-asserted-by":"crossref","first-page":"682","DOI":"10.1108\/DTA-10-2020-0239","article-title":"Attack detection in medical internet of things using optimized deep learning: enhanced security in healthcare sector","volume":"55","year":"2021","journal-title":"Data Technologies and Applications"},{"issue":"11","key":"key2023071009261008200_ref018","doi-asserted-by":"crossref","first-page":"9549","DOI":"10.1609\/aaai.v35i11.17150","article-title":"Membership privacy for machine learning models through knowledge transfer","volume":"35","year":"2021","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"first-page":"3","article-title":"Membership inference attacks against machine learning models","year":"2017","key":"key2023071009261008200_ref019"},{"year":"2014","key":"key2023071009261008200_ref020","article-title":"Very deep convolutional networks for large-scale image recognition"},{"key":"key2023071009261008200_ref021","doi-asserted-by":"crossref","first-page":"115","DOI":"10.1016\/j.inffus.2019.12.001","article-title":"A survey on machine learning for data fusion","volume":"57","year":"2020","journal-title":"Information Fusion"},{"year":"2010","key":"key2023071009261008200_ref022","article-title":"Caltech-UCSD birds 200"},{"first-page":"268","article-title":"Privacy risk in machine learning: analyzing the connection to overfitting","year":"2018","key":"key2023071009261008200_ref023"}],"container-title":["International Journal of Web Information Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.emerald.com\/insight\/content\/doi\/10.1108\/IJWIS-03-2023-0050\/full\/xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.emerald.com\/insight\/content\/doi\/10.1108\/IJWIS-03-2023-0050\/full\/html","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,24]],"date-time":"2025-07-24T22:23:53Z","timestamp":1753395833000},"score":1,"resource":{"primary":{"URL":"http:\/\/www.emerald.com\/ijwis\/article\/19\/2\/61-79\/160721"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,7,10]]},"references-count":23,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2023,7,10]]},"published-print":{"date-parts":[[2023,7,12]]}},"alternative-id":["10.1108\/IJWIS-03-2023-0050"],"URL":"https:\/\/doi.org\/10.1108\/ijwis-03-2023-0050","relation":{},"ISSN":["1744-0084","1744-0084"],"issn-type":[{"type":"print","value":"1744-0084"},{"type":"electronic","value":"1744-0084"}],"subject":[],"published":{"date-parts":[[2023,7,10]]}}}