8,797 Hits in 3.9 sec

Interpretable Machine Learning for Privacy-Preserving Pervasive Systems [article]

Benjamin Baron, Mirco Musolesi
2019 arXiv   pre-print
In this paper, we propose a machine learning interpretability framework that enables users to understand how these generated traces violate their privacy.  ...  Our everyday interactions with pervasive systems generate traces that capture various aspects of human behavior and enable machine learning algorithms to extract latent information about users.  ...  Figure 1 . 1 Interpretable Machine Learning for Privacy-Preserving Pervasive Systems. (1) Our client is installed on the user's device.  ... 
arXiv:1710.08464v6 fatcat:fv66extdtzf65ofz7amyjwhdqq

Societal Issues in Machine Learning: When Learning from Data is Not Enough

Davide Bacciu, Battista Biggio, Paulo Lisboa, José D. Martín, Luca Oneto, Alfredo Vellido
2019 The European Symposium on Artificial Neural Networks  
This phenomenon means that AI systems and products are reaching the society at large and, therefore, that societal issues related to the use of AI and Machine Learning (ML) cannot be ignored any longer  ...  Designing ML models from this human-centered perspective means incorporating human-relevant requirements such as safety, fairness, privacy, and interpretability, but also considering broad societal issues  ...  One way to preserve privacy is to corrupt the learning procedure with noise without destroying the information that we want to extract.  ... 
dblp:conf/esann/BacciuBLMOV19 fatcat:ltuya4xxvbgnpm67p6zikh34ry

Digital Divide and Social Dilemma of Privacy Preservation [article]

Hamoud Alhazmi, Ahmed Imran, Mohammad Abu Alsheikh
2021 arXiv   pre-print
For example, the various levels of government legislation and compliance on information privacy worldwide have created a new era of digital divide in the privacy preservation domain.  ...  In this article, the concept "digital privacy divide (DPD)" is introduced to describe the perceived gap in the privacy preservation of individuals based on the geopolitical location of different countries  ...  DPD in machine learning: Machine learning, e.g., deep learning, is widely adopted for extracting hidden patterns from the collected data.  ... 
arXiv:2110.02669v1 fatcat:plwnujgbtrhtfik5ezkjbopfae

Pervasive computing at scale: Transforming the state of the art

Diane J. Cook, Sajal K. Das
2012 Pervasive and Mobile Computing  
computing systems and services with diverse applications and global accessibility.  ...  In this paper we assess the current state of the art in of pervasive computing at scale (PeCS) and look ahead to future directions the field can pursue together with challenges it will need to overcome  ...  Acknowledgements We would like to acknowledge the roles of Andrew Campbell, Roy Want, and Shwetak Patel in organizing (along with the authors of this article) the NSF PeCS workshop and thank them for their  ... 
doi:10.1016/j.pmcj.2011.10.004 fatcat:quu5gtg3h5fvjcyxg44mnmjk5i

Making machine learning trustworthy

Birhanu Eshete
2021 Science  
Issues include COMPUTER SCIENCE Making machine learning trustworthy Safety, transparency, and fairness are essential for high-stakes uses of machine learning Training data and model under attack  ...  Machine learning models can be victims of malicious attacks during training and deployment.  ... 
doi:10.1126/science.abi5052 fatcat:qjnee5ile5ftbbdgwvkh65dima

A semantic approach for building pervasive spaces

Daniel Massaguer, Sharad Mehrotra, Nalini Venkatasubramanian
2009 Proceedings of the 6th Middleware Doctoral Symposium on - MDS '09  
Large and pervasive sensing, communications, and computing infrastructures are enabling the realization of pervasive spaces. Enabling such spaces, however, encompasses a set of challenges.  ...  to applications, and (iii) a set of mechanisms that are able to protect privacy due to the inclusion of semantics in the middleware.  ...  Their valuable input and their contributions on coauthoring papers and code has made it possible for this research to come to live.  ... 
doi:10.1145/1659753.1659755 fatcat:zevs3fpkpvdtfikgf3ghbmifvm

Is There an App for That?

M. Langheinrich
2020 IEEE pervasive computing  
And what does that mean for smart wearables, such as smart watches and smart glasses? 1536-1268 ß 2020 IEEE Published by the IEEE Computer Society  ...  The second feature article by Baron and Musolesi proposed a novel framework for "Interpretable Machine Learning for Privacy-Preserving Pervasive Systems."  ...  Simon holds the Chair for Interaction-and Communication-Based Systems at the University of St. Gallen, Switzerland.  ... 
doi:10.1109/mprv.2020.2969587 fatcat:pgwgh4tcknabpajqzrgbu5o5xu

MUSKETEER D2.3 Key performance indicators selection and definition

Susanna Bonura, Davide Dalle Carbonare
2020 Zenodo  
Detailed description of the technical and domain business-specific KPIs that will be used for validating the MUSKETEER data platform.  ...  The Evaluation Framework employed for validating the MUSKETEER platform, is based on the Goal Question Metric (GQM) method.  ...  Machine Learning over a high variety of different privacy-preserving scenarios. O1.1.  ... 
doi:10.5281/zenodo.4730055 fatcat:qksagw5sffasnn7xairvylyqm4

D2.7 Key performance indicators selection and definition - final version

Susanna Bonura, Davide Dalle Carbonare
2020 Zenodo  
The Evaluation Framework employed for validating the MUSKETEER platform, is based on the Goal Question Metric (GQM) method.  ...  Starting from this final version of the MUSKETEER evaluation framework and KPIs, data collection and interpretation phases will be documented in the deliverables D7.5 and D7.6, where the description of  ...  Machine Learning to Augment Shared Knowledge in Federated Privacy-Preserving Scenarios (MUSKETEER) In terms of input data for system training reasons, Hygeia will draw all pelvis MRI exams as well as multi-parametric  ... 
doi:10.5281/zenodo.5845684 fatcat:zedfi6jdibgptbherd6rp2abte

Automated activity recognition and monitoring of elderly using wireless sensors: Research challenges

Damith C. Ranasinghe, Roberto L. Shinmoto Torres, Asanga Wickramasinghe
2013 5th IEEE International Workshop on Advances in Sensors and Interfaces IWASI  
This leads to the requirement for more advanced machine learning paradigms [16] .  ...  Thus, researchers are lured into the machine learning domain in learning generalized classification models.  ... 
doi:10.1109/iwasi.2013.6576067 dblp:conf/iwasi/RanasingheTW13 fatcat:zes6c5ibynfctkodv2yurrklii

Highlights Society Magazines

2021 Computer  
Society's lineup of 12 peer-reviewed technical magazines covers cutting-edge topics ranging from software design and computer graphics to Internet computing and security, from scientific applications and machine  ...  To facilitate this, privacy-preserving solutions are in high demand.  ...  In this article from the January/February 2021 issue of IEEE Intelligent Systems, the authors develop a distributed collaborative coupling learning system, which enables differential privacy and defends  ... 
doi:10.1109/mc.2021.3055875 fatcat:jrvp2ke2sfgxzlkd3gprevir54

Session Introduction

Steven E. Brenner, Martha L. Bulyk, Dana C. Crawford, Alexander A. Morgan, Predrag Radivojac, Nicholas P. Tatonetti
2020 Pacific Symposium on Biocomputing  
Yet to bridge molecular measurement, clinical observation, social interaction and clinical practice, substantial gains must be made in methods for data integration, analysis, model development, and interpretation  ...  These present daunting challenges of how to ethically share and analyze data, as we aim to ensure privacy, personal initiative, and mitigate disparities.  ...  of outputs of machine learning tools.  ... 
dblp:conf/psb/BrennerBCMRT20 fatcat:faa3tncfszbgdaslqx5pwi4o3e

Middleware for Pervasive Spaces: Balancing Privacy and Utility [chapter]

Daniel Massaguer, Bijit Hore, Mamadou H. Diallo, Sharad Mehrotra, Nalini Venkatasubramanian
2009 Lecture Notes in Computer Science  
Middleware for pervasive spaces has to meet conflicting requirements. It has to both maximize the utility of the information exposed and ensure that this information does not violate users' privacy.  ...  We begin by providing appropriate definitions of privacy and utility for the type of applications that would support collaborative work in an office environment-current definitions of privacy and anonymity  ...  Acknowledgements The authors would like to thank the SATware team for their dedication to the project and specially to Roberto Gamboni and Jay Lickfett for his help on mantaining and extending the implementation  ... 
doi:10.1007/978-3-642-10445-9_13 fatcat:iszcsu52lncp7mnahrj3ghvyhy

Privacy Issues in Holistic Recommendations

Federica Cena, Ruggero G. Pensa, Amon Rapp
2019 Adjunct Publication of the 27th Conference on User Modeling, Adaptation and Personalization - UMAP'19 Adjunct  
CCS CONCEPTS • Human-centered computing → HCI theory, concepts and models; • Security and privacyPrivacy protections.  ...  In this paper we point out some relevant issues in relation to privacy when providing holistic recommendations.  ...  Privacy preservation.  ... 
doi:10.1145/3314183.3323461 dblp:conf/um/CenaPR19 fatcat:uqjwsrkuujh7tj3ba47fmqutju

Solving the Challenges of Pervasive Computing

Syed Muqsit Shaheed, Jalil Abbas, Asif Shabbir, Fayyaz Khalid
2015 Journal of Computer and Communications  
Pervasive systems have a broad range of applications but it is relatively challenging for pervasive applications to meet emergence into existing physical environment and newly built structure requirements  ...  Modern research emphasizes Pervasive Computing change faces, learning cultures, structures, communications, intellectual properties, information securities, data presentations and web displays to make  ...  Interaction in the middle of human and machine brings consideration regarding, particular explanatory parts of interaction design are availability, interpretability and connectivity [4] .  ... 
doi:10.4236/jcc.2015.39005 fatcat:auqkzu5ivveglaeqthm4vzoz7e
« Previous Showing results 1 — 15 out of 8,797 results