Respecting Partial Privacy of Unstructured Data via Spectrum-Based Encoder
Since the popularity of Machine Learning as a Service (MLaaS) has been increasing significantly, users are facing the risk of exposing sensitive information that is not task-related. The reason is that the data uploaded by users may include some information that is not useful for inference but can l...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2024-02-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/24/3/1015 |
_version_ | 1797318215457570816 |
---|---|
author | Qingcai Luo Hui Li |
author_facet | Qingcai Luo Hui Li |
author_sort | Qingcai Luo |
collection | DOAJ |
description | Since the popularity of Machine Learning as a Service (MLaaS) has been increasing significantly, users are facing the risk of exposing sensitive information that is not task-related. The reason is that the data uploaded by users may include some information that is not useful for inference but can lead to privacy leakage. One straightforward approach to mitigate this issue is to filter out task-independent information to protect user privacy. However, this method is feasible for structured data with naturally independent entries, but it is challenging for unstructured data. Therefore, we propose a novel framework, which employs a spectrum-based encoder to transform unstructured data into the latent space and a task-specific model to identify the essential information for the target task. Our system has been comprehensively evaluated on three benchmark visual datasets and compared to previous works. The results demonstrate that our framework offers superior protection for task-independent information and maintains the usefulness of task-related information. |
first_indexed | 2024-03-08T03:49:11Z |
format | Article |
id | doaj.art-cdc0ce2860594e02b6bff77c6d963513 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-08T03:49:11Z |
publishDate | 2024-02-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-cdc0ce2860594e02b6bff77c6d9635132024-02-09T15:22:35ZengMDPI AGSensors1424-82202024-02-01243101510.3390/s24031015Respecting Partial Privacy of Unstructured Data via Spectrum-Based EncoderQingcai Luo0Hui Li1School of Cyber Engineering, Xidian University, Xi’an 710126, ChinaSchool of Computer Science and Technology, Xidian University, Xi’an 710071, ChinaSince the popularity of Machine Learning as a Service (MLaaS) has been increasing significantly, users are facing the risk of exposing sensitive information that is not task-related. The reason is that the data uploaded by users may include some information that is not useful for inference but can lead to privacy leakage. One straightforward approach to mitigate this issue is to filter out task-independent information to protect user privacy. However, this method is feasible for structured data with naturally independent entries, but it is challenging for unstructured data. Therefore, we propose a novel framework, which employs a spectrum-based encoder to transform unstructured data into the latent space and a task-specific model to identify the essential information for the target task. Our system has been comprehensively evaluated on three benchmark visual datasets and compared to previous works. The results demonstrate that our framework offers superior protection for task-independent information and maintains the usefulness of task-related information.https://www.mdpi.com/1424-8220/24/3/1015spectrum-based encoderlatent codemachine learningprivacy preserving |
spellingShingle | Qingcai Luo Hui Li Respecting Partial Privacy of Unstructured Data via Spectrum-Based Encoder Sensors spectrum-based encoder latent code machine learning privacy preserving |
title | Respecting Partial Privacy of Unstructured Data via Spectrum-Based Encoder |
title_full | Respecting Partial Privacy of Unstructured Data via Spectrum-Based Encoder |
title_fullStr | Respecting Partial Privacy of Unstructured Data via Spectrum-Based Encoder |
title_full_unstemmed | Respecting Partial Privacy of Unstructured Data via Spectrum-Based Encoder |
title_short | Respecting Partial Privacy of Unstructured Data via Spectrum-Based Encoder |
title_sort | respecting partial privacy of unstructured data via spectrum based encoder |
topic | spectrum-based encoder latent code machine learning privacy preserving |
url | https://www.mdpi.com/1424-8220/24/3/1015 |
work_keys_str_mv | AT qingcailuo respectingpartialprivacyofunstructureddataviaspectrumbasedencoder AT huili respectingpartialprivacyofunstructureddataviaspectrumbasedencoder |