Protecting Sensory Data against Sensitive Inferences
Open accessThe DOI matches its Crossref record; 14 of the 14 references it cites were found in a registry or an index.
Abstract
There is growing concern about how personal data are used when users grant applications direct access to the sensors of their mobile devices. In fact, high resolution temporal data generated by motion sensors reflect directly the activities of a user and indirectly physical and demographic attributes. In this paper, we propose a feature learning architecture for mobile devices that provides flexible and negotiable privacy-preserving sensor data transmission by appropriately transforming raw sensor data. The objective is to move from the current binary setting of granting or not permission to an application, toward a model that allows users to grant each application permission over a limited range of inferences according to the provided services. The internal structure of each component of the proposed architecture can be flexibly changed and the trade-off between privacy and utility can be negotiated between the constraints of the user and the underlying application. We validated the proposed architecture in an activity recognition application using two real-world datasets, with the objective of recognizing an activity without disclosing gender as an example of private information. Results show that the proposed framework maintains the usefulness of the transformed data for activity recognition, with an average loss of only around three percentage points, while reducing the possibility of gender classification to around 50%, the target random guess, from more than 90% when using raw sensor data. We also present and distribute MotionSense, a new dataset for activity and attribute recognition collected from motion sensors.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“This dataset consists of accelerometer, gyroscope, and altitude data collected with an iPhone 6s using SensingKit with a sampling rate of 50 Hz . 24 participants of different gender, age, weight, and height groups have performed 6 activities (walking downstairs, walking upstairs, walking, sitting, standing, and jogging) in 15 trials, while carrying the smartphone in their trousers' front pocket.”
From Self-supervised Human Activity Recognition by Learning to Predict Cross-Dimensional Motion (2021) · cites this paper as [13]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Experimental results demonstrate that the proposed framework can maintain the usefulness of the transformed data for activity recognition while reducing the gender classification accuracy to 50% (random guessing) from more than 90% when using raw sensor data.”
From Deep Learning in Human Activity Recognition with Wearable Sensors: A Review on Advances (2022) · cites this paper as [128]
The cited source says what the claim attributes to it.
In the citing paper
“Multiple works have proposed privacy preservation methods while classifying human activities, including the replacement auto-encoder, the guardian, estimator, and neutralizer (GEN) architecture [128], and the anonymizing autoencoder [129].”
From Deep Learning in Human Activity Recognition with Wearable Sensors: A Review on Advances (2022) · cites this paper as [128]
The cited source says what the claim attributes to it.
In the citing paper
“Three publicly available datasets, UCI HAR [1], MotionSense [13], and HAPT [18] are used to evaluate our model.”
From Self-supervised Human Activity Recognition by Learning to Predict Cross-Dimensional Motion (2021) · cites this paper as [13]
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“as it is more computationally efficient while also being comparable in performance to grid search [44].”
From Investigating Enhancements to Contrastive Predictive Coding for Human Activity Recognition (2023) · cites this paper as [44]
Experimental — accuracy not yet measured
6 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Extracting and composing robust features with denoising autoencoders
Vincent, P., Larochelle, H., Bengio, Y., & Manzagol, P.-A. (2008) · Proceedings of the 25th international conference on Machine learning - ICML '08 · doi:10.1145/1390156.1390294 · Cited by 7,436 · cited 1×
“In the specific implementation of this paper, we use a deep autoencoder [16] as Guardian.”
MethodDifferential Privacy: A Survey of Results
Dwork, C. (2008) · Lecture Notes in Computer Science · doi:10.1007/978-3-540-79228-4_1 · Cited by 3,542 · cited 1×
“To provide a formal privacy guarantee, [2] trains GANs under the constraint of differential privacy [5] to protect against common privacy attacks.”
BackgroundBagnall, A., Lines, J., Bostrom, A., Large, J., & Keogh, E. (2017) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-016-0483-9 · Cited by 1,377
Generating Multi-label Discrete Patient Records using Generative Adversarial Networks
Choi, E., Biswal, S., Malin, B., Duke, J. D., Stewart, W. F., & Sun, J. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1703.06490 · Cited by 341 · cited 1×
“For example, to protect health records synthetic medical datasets can be published instead of the real ones using generative models training on sensitive real-world medical datasets [3, 6].”
BackgroundThe MobiAct Dataset: Recognition of Activities of Daily Living using Smartphones
Vavoulas, G., Chatzaki, C., Malliotakis, T., Pediaditis, M., & Tsiknakis, M. (2016) · Proceedings of the International Conference on Information and Communication Technologies for Ageing Well and e-Health · doi:10.5220/0005792401430151 · Cited by 301 · cited 1×
“MobiAct [15] includes sensor data (accelerometer, gyroscope, orientation) from a smartphone collected when data subjects performed nine activities and a range of falls.”
MethodReal-valued (Medical) Time Series Generation with Recurrent Conditional GANs
Esteban, C., Hyland, S. L., & Rätsch, G. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1706.02633 · Cited by 253 · cited 1×
“For example, to protect health records synthetic medical datasets can be published instead of the real ones using generative models training on sensitive real-world medical datasets [3, 6].”
BackgroundContext-Aware Generative Adversarial Privacy
Huang, C., Kairouz, P., Chen, X., Sankar, L., & Rajagopal, R. (2017) · Entropy · doi:10.3390/e19120656 · Cited by 156
Personal Data: Thinking Inside the Box
Chaudhry, A., Crowcroft, J., Howard, H., Madhavapeddy, A., Mortier, R., Haddadi, H., & McAuley, D. (2015) · Aarhus Series on Human Centered Computing · doi:10.7146/aahcc.v1i1.21312 · Cited by 139
Self-citationToward accurate dynamic time warping in linear time and space
SalvadorStan, & ChanPhilip (2007) · doi:10.5555/1367985.1367993 · Cited by 103
Privacy-Preserving Adversarial Networks
Tripathy, A., Wang, Y., & Ishwar, P. (2019) · 2019 57th Annual Allerton Conference on Communication, Control, and Computing (Allerton) · doi:10.1109/allerton.2019.8919758 · Cited by 87 · cited 1×
“GANs have also been applied for enhancing privacy [9, 14].”
Method
Research works printed but not linked: 4
- Beaulieu-Jones B. K., 2017, Privacy-preserving generative deep neural networks support clinical data sharing. bioRxiv, page 159756
- 10.5555/1791834.1791836 · doi:10.5555/1791834.1791836
- Goodfellow I., 2014, Advances in neural information processing systems, 2672
- 10.5555/2832747.2832806 · doi:10.5555/2832747.2832806
Cited by
Deep Learning for Sensor-based Human Activity Recognition
Chen, K., Zhang, D., Yao, L., Guo, B., Yu, Z., & Liu, Y. (2022) · ACM Computing Surveys · doi:10.1145/3447744 · Cited by 690
“rocess. For example, Iwasawa et al. [65] proposed to integrate an adversarial loss with the standard activity classification loss to minimize the user identification accuracy. The authors of [92] and [91] also adopted the similar idea to prevent privacy leakage. Their experiment results show an effective reduction of…”
BackgroundDeep Learning in Human Activity Recognition with Wearable Sensors: A Review on Advances
Zhang, S., Li, Y., Zhang, S., Shahabi, F., Xia, S., Deng, Y., & Alshurafa, N. (2022) · Sensors · doi:10.3390/s22041476 · Cited by 625
“introduces a framework called Guardian-Estimator-Neutralizer (GEN) that attempts to recognize activities while preserving gender privacy [180].”
MethodMulti-task Self-Supervised Learning for Human Activity Detection
Saeed, A., Ozcelebi, T., & Lukkien, J. (2019) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3328932 · Cited by 346
“e MotionSense dataset [40] comprises an accelerometer, gyroscope, and altitude data from 24 participants of varying age, gender, weight, and height.”
BackgroundDeep Learning for Sensor-based Human Activity Recognition: Overview, Challenges and Opportunities
Chen, K., Zhang, D., Yao, L., Guo, B., Yu, Z., & Liu, Y. (2020) · arXiv · doi:10.48550/arxiv.2001.07416 · Cited by 314
“rocess. For example, Iwasawa et al. [65] proposed to integrate an adversarial loss with the standard activity classification loss to minimize the user identification accuracy. The authors of [92] and [91] also adopted the similar idea to prevent privacy leakage. Their experiment results show an effective reduction of…”
BackgroundMobile sensor data anonymization
Malekzadeh, M., Clegg, R. G., Cavallaro, A., & Haddadi, H. (2019) · Proceedings of the International Conference on Internet of Things Design and Implementation · doi:10.1145/3302505.3310068 · Cited by 298
Self-citation
Metadata
- Authors
- Malekzadeh, M., Clegg, R. G., Cavallaro, A., & Haddadi, H.
- Affiliations
- Queen Mary University of London (GB)
- Imperial College London (GB)
- Venue
- Proceedings of the 1st Workshop on Privacy by Design in Distributed Systems
- Publisher
- ACM
- Published
- 2018-04-23
- Pages
- 1-6
- OpenAlex
- W2789136377
- Cited by
- 154
- References
- 9
- Citations by year
- 2026: 6 · 2025: 25 · 2024: 38 · 2023: 25 · 2022: 14 · 2021: 22 · 2020: 15 · 2019: 8 · 2018: 1
- FWCI
- 4.68
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Queen Mary University of London Life Science Initiative
- Microsoft Azure for Research Award
- EPSRC Databox grant (EP/N028260/1)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Social Sciences › Sociology and Political Science › Privacy, Security, and Data Protection
- Keywords
- mobile device sensors, representation learning, privacy-utility trade-off, gender inference, activity recognition, motion sensors, sensitive attribute inference
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Protecting Sensory Data against Sensitive Inferences. https://app.cochat.ai/papers/10.1145/3195258.3195260
- The paper
- Malekzadeh, M., Clegg, R. G., Cavallaro, A., & Haddadi, H. (2018). Protecting Sensory Data against Sensitive Inferences. Proceedings of the 1st Workshop on Privacy by Design in Distributed Systems. https://doi.org/10.1145/3195258.3195260
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.