Federated K-Means Clustering: A Novel Edge AI Based Approach for Privacy Preservation

Kumar, Hemant; Karthik, V R; Nair, Mydhili K.

doi:10.1109/ccem50674.2020.00021

Cited by 17 publications

(5 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are also distributed versions of this algorithm. The Federated Learning version of k-means combines ideas from the distributed algorithm with the concept of only using a fraction of the clients on each iteration [23]. Our implementation of federated k-means is an adaptation of the ideas presented in [24].…”

Section: Task 2: Unsupervised Student Classificationmentioning

confidence: 99%

Federated Learning for Education Data Analytics

Fachola¹,

Tornaria²,

Bermolen³

et al. 2023

Preprint

View full text Add to dashboard Cite

Federated learning techniques aim to train and build machine learning models based on distributed datasets across multiple devices, avoiding data leakage. The main idea is to perform training on remote devices or isolated data centers without transferring data to centralized repositories, thus mitigating privacy risks. Data analytics in education, in particular learning analytics, is a promising scenario to apply this approach to address the legal and ethical issues related to processing sensitive data. Indeed, given the nature of the data to be studied (personal data, educational outcomes, data concerning minors), it is essential to ensure that the conduct of these studies and the publication of the results provide the necessary guarantees to protect the privacy of the individuals involved and the protection of their data. In addition, the application of quantitative techniques based on the exploitation of data on the use of educational platforms, student performance, use of devices, etc., can account for educational problems such as the determination of user profiles, personalized learning trajectories, or early dropout indicators and alerts, among others. This paper presents the application of federated learning techniques to two learning analytics problems: dropout prediction and unsupervised student classification. The experiments allow us to conclude that the proposed solutions achieve comparable results from the performance point of view with the centralized versions, avoiding centralizing the data for training the models.

show abstract

Section: Task 2: Unsupervised Student Classificationmentioning

confidence: 99%

Federated Learning for Education Data Analytics

Fachola¹,

Tornaria²,

Bermolen³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Sachdev presented security and privacy issues of Edge AI in digital marketing and concluded that one of the main challenges is how Edge AI can extensively be implemented in that context [6]. Kumar et al proved by using the classical k-means algorithm in Edge AI concept the feasibility of maintaining privacy preservation of data with Edge AI processing [7].…”

Section: Introductionmentioning

confidence: 99%

Artificial Intelligence in the IoT Era: A Review of Edge AI Hardware and Software

Sipola

Alatalo

Kokkonen

et al. 2022

2022 31st Conference of Open Innovations Association (FRUCT)

View full text Add to dashboard Cite

The modern trend of moving artificial intelligence computation near to the origin of data sources has increased the demand for new hardware and software suitable for such environments. We carried out a scoping study to find the current resources used when developing Edge AI applications. Due to the nature of the topic, the research combined scientific sources with product information and software project sources. The paper is structured as follows. In the first part, Edge AI applications are briefly discussed followed by hardware options and finally, the software used to develop AI models is described. There are various hardware products available, and we found as many as possible for this research to identify the best-known manufacturers. We describe the devices in the following categories: artificial intelligence accelerators and processors, field-programmable gate arrays, system-on-a-chip devices, system-on-modules, and full computers from development boards to servers. There seem to be three trends in Edge AI software development: neural network optimization, mobile device software and microcontroller software. We discussed these emerging fields and how the special challenges of low power consumption and machine learning computation are being taken into account. Our findings suggest that the Edge AI ecosystem is currently developing, and it has its own challenges to which vendors and developers are responding.

show abstract

“…This technique is helpful for B5G/6G networks because, With the ever-increasing edge connectivity, it will create huge network traffic from billions of these small devices if a central server performs AI computations on their data. Therefore, to mitigate this issue, it is inevitable to bring AI functions to the edge by offloading intelligence from the cloud to these devices to facilitate "bringing code to data, not data to code" [198].…”

Section: Edge Ai 1) Introductionmentioning

confidence: 99%

“…Another privacy preserved implementation of edge AI is in [200], where the authors propose a privacy-preserving AI task composition framework for pushing AI tasks to the edge networks based on homomorphic encryption. [198] suggests a federated k-means clustering for edge networks with privacy preservation since the data does not leave the edge device.…”

Section: Edge Ai 1) Introductionmentioning

confidence: 99%

A Survey on Privacy for B5G/6G: New Privacy Challenges, and Research Directions

Sandeepa¹,

Siniarski²,

Kourtellis³

et al. 2022

Preprint

View full text Add to dashboard Cite

Massive progress of mobile wireless telecommunication networks was achieved in the previous decades, with privacy enhancement in each. At present, mobile users are getting familiar with the latest 5G networks, and the discussion for the next generation of Beyond 5G (B5G)/6G networks has already been initiated. It is expected that B5G/6G will push the existing network capabilities to the next level, with higher speeds, enhanced reliability, and seamless connectivity. To make these expectations a reality, research is progressing on new technologies, architectures, and intelligence-based decision-making processes related to B5G/6G. Privacy considerations are a crucial aspect that needs further attention in such developments, as billions of people and devices will be transmitting their data through the upcoming network. This paper provides a comprehensive survey on privacy-related aspects for B5G/6G networks. First, it discusses a taxonomy of different privacy perspectives. Based on the taxonomy, the paper then conceptualizes a set of privacy goals for the B5G/6G and the challenges that appear as barriers to reaching these goals. Next, this work provides a set of solutions applicable to the proposed architecture of B5G/6G networks to mitigate the challenges. Additionally, this paper discusses the emerging field of non-personal data privacy. It also provides an overview of standardization initiatives for privacy preservation. Finally, it concludes with a roadmap of future directions and upcoming trends containing privacy-related topics, which will be an arena for new research towards privacy-enhanced B5G/6G networks. This work provides a basis for privacy aspects that will significantly impact peoples' daily lives with future networks.

show abstract

Federated K-Means Clustering: A Novel Edge AI Based Approach for Privacy Preservation

Cited by 17 publications

References 3 publications

Federated Learning for Education Data Analytics

Federated Learning for Education Data Analytics

Artificial Intelligence in the IoT Era: A Review of Edge AI Hardware and Software

A Survey on Privacy for B5G/6G: New Privacy Challenges, and Research Directions

Contact Info

Product

Resources

About