Efficient Online Decentralized Learning Framework for Social Internet of Things

Ching, Cheng-Wei; Huang, Hung-Sheng; Yang, Chun-An; Kuo, Jian-Jhih; Hwang, Ren-Hung

doi:10.1109/globecom46510.2021.9685824

Cited by 3 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To mitigate the possible communication and computational bottleneck, we device the decentralized serverless model fine-tuning. Specifically, the leaf nodes first perform local model fine-tuning and then upload their gradients to one of leaf nodes that have social links one another [13,17]. After receiving the intermediate fine-tuned models from their trust leaf nodes, the leaf nodes aggregate the intermediate fine-tuned models and continue uploading gradients until the root receives the intermediate fine-tuned models.…”

Section: Online Model Fine-tuning and Ensemble-based Model Inferencementioning

confidence: 99%

Decaffe: DHT Tree-Based Online Federated Fake News Detection

Ching,

2024

2024 8th International Conference on Control Engineering and Artificial Intelligence

View full text Add to dashboard Cite

The proliferation of mobile social networks (MSNs) has transformed information dissemination, leading to increased reliance on these platforms for news consumption. However, this shift has been accompanied by the widespread propagation of fake news, posing significant challenges in terms of public panic, political influence, and the obscuring of truth. Traditional data processing pipelines for fake news detection in MSNs suffer from lengthy response times and poor scalability, failing to address the unique characteristics of news in MSNs, such as prompt propagation, large-scale quantity, and rapid evolution. This paper introduces a novel system named Decaffe -a DHT Tree-Based Online Federated Fake News Detection system. Decaffe leverages distributed hash table (DHT)-based aggregation trees for scalability and real-time detection, and it employs two model fine-tuning methods for adapting to mobile network dynamics. The system's structure includes a root, branches, and leaves for effective dissemination of a pre-trained model and ensemblebased aggregation of predictive results. Decaffe uniquely combines centralized server-based and decentralized serverless model finetuning approaches with personalized model fine-tuning, addressing the challenges of real-time detection, scalability, and adaptability in the dynamic environment of MSNs. CCS Concepts• Computer systems organization → Peer-to-peer architectures.

show abstract

Section: Online Model Fine-tuning and Ensemble-based Model Inferencementioning

confidence: 99%

Decaffe: DHT Tree-Based Online Federated Fake News Detection

Ching,

2024

2024 8th International Conference on Control Engineering and Artificial Intelligence

View full text Add to dashboard Cite

show abstract

“…The suggested model from (Mohammadi et al, 2021) was then tested using the Netlogo simulator and compared to various scale-free architectures such as large components, Barab'asi-Albert networks and Random networks. The designed framework identifies each device's neighbor and constructs a communication topology based on a social network (Ching et al, 2021). Ramasamy and Arjunasamy (2017) suggested metrices to be used while selecting objects friendship links are (1) Number of friends node harm.…”

Section: Friend Selectionmentioning

confidence: 99%

A Comprehensive Survey on Security, Trust Management and Privacy Preservation for Social-Internet-of-Things (S-IoT)

Rahul¹,

Venkatesh²,

Venugopal³

2023

Journal of Computer Science

View full text Add to dashboard Cite

The Social Internet of Things (S-IoT) is a new paradigm that evolved with a convergence of social networking concepts and internet of things concepts. This paradigm helps develop intelligent devices and services with higher utility. The new paradigm enriches the social interaction between human's objects, heterogenous social objects interaction, the discovery of service and scalability. In recent years, S-IoT has been researching a hot topic and has drawn the attention of researchers, resulting in research on trust management, privacy preservation, service provisioning and security mechanisms. However, S-IoT faces new research challenges like interoperability, trustworthiness, security threats, scalability, navigability, service discovery and composition. In line with the research trend, this article provides a systematic survey of fundamental concepts of S-IoT and IoT, distinguishable characteristics of S-IoT and IoT and research challenges in S-IoT. This survey thoroughly reviews secure communication mechanisms, privacy-preservation techniques and the trust-evaluation model in S-IoT. Furthermore, we accentuate the privacy-preservation mechanism proposed in the literature for seven years and explored research areas. The holistic survey gives an overview of trust-evaluation models with respect to the service provider's perspective. This article overviews secure communication protocols and architecture in the S-IoT context. Finally, various research gaps, applications of S-IoT and research directions help a researcher to carry out research work in S-IoT.

show abstract