{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/replacement-autoencoder-a-privacy-preserving","title":"Replacement AutoEncoder: A Privacy-Preserving Algorithm for Sensory Data Analysis","arxiv_id":"1710.06564","date":"2017-10-18","proceeding":null,"authors":["Mohammad Malekzadeh","Richard G. Clegg","Hamed Haddadi"],"abstract":"An increasing number of sensors on mobile, Internet of things (IoT), and\nwearable devices generate time-series measurements of physical activities.\nThough access to the sensory data is critical to the success of many beneficial\napplications such as health monitoring or activity recognition, a wide range of\npotentially sensitive information about the individuals can also be discovered\nthrough access to sensory data and this cannot easily be protected using\ntraditional privacy approaches.\n  In this paper, we propose a privacy-preserving sensing framework for managing\naccess to time-series data in order to provide utility while protecting\nindividuals' privacy. We introduce Replacement AutoEncoder, a novel algorithm\nwhich learns how to transform discriminative features of data that correspond\nto sensitive inferences, into some features that have been more observed in\nnon-sensitive inferences, to protect users' privacy. This efficiency is\nachieved by defining a user-customized objective function for deep\nautoencoders. Our replacement method will not only eliminate the possibility of\nrecognizing sensitive inferences, it also eliminates the possibility of\ndetecting the occurrence of them. That is the main weakness of other approaches\nsuch as filtering or randomization. We evaluate the efficacy of the algorithm\nwith an activity recognition task in a multi-sensing environment using\nextensive experiments on three benchmark datasets. We show that it can retain\nthe recognition accuracy of state-of-the-art techniques while simultaneously\npreserving the privacy of sensitive information. Finally, we utilize the GANs\nfor detecting the occurrence of replacement, after releasing data, and show\nthat this can be done only if the adversarial network is trained on the users'\noriginal data.","url_abs":"http://arxiv.org/abs/1710.06564v3","url_pdf":"http://arxiv.org/pdf/1710.06564v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"replacement-autoencoder-a-privacy-preserving","repo_url":"https://github.com/mmalekzadeh/replacement-autoencoder","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"activity-recognition","task_name":"Activity Recognition"},{"task_slug":"privacy-preserving","task_name":"Privacy Preserving"},{"task_slug":"time-series-1","task_name":"Time Series"},{"task_slug":"time-series","task_name":"Time Series Analysis"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}