Nov 2026· IEEE Internet of Things Journal· Vol 13, pp. 53165-53178· 0 citations· 65 references
TL;DR
This work introduces a two-stage source localization with subgraph GNNs (TSLSGs) framework and employs subgraph GNNs to restrict the message-passing receptive field strictly within localized subgraphs, effectively preserving local heterogeneity and mitigating the oversmoothing problem.
Abstract
With the advent of the Internet of Things (IoT) era, a large number of sensors have been deployed to obtain node states, bringing the source-localization task—identifying the locations of multiple unknown propagation sources using sensory data—into sharp focus. Rapidly and accurately identifying the sources of misinformation or disease outbreaks is crucial for managing and controlling harmful propagation processes. While graph neural networks (GNNs) have been widely adopted to aggregate infection information from neighboring nodes, existing GNN-based methods generally face two practical limitations. First, they simulate the propagation process relying on network topologies, failing to accurately capture authentic propagation dynamics. Second, the inherent risk of oversmoothing during multihop neighborhood aggregation weakens local heterogeneity and obscures explicit topological signatures, such as differences in source centrality. To address these issues, we introduce a two-stage source localization with subgraph GNNs (TSLSGs) framework. In particular, in the first stage, we train a propagation prediction network using actual subsequent infection states. This stage accurately captures genuine propagation dynamics, thereby inherently magnifying source centrality and filtering out observational noise. Furthermore, we employ subgraph GNNs to restrict the message-passing receptive field strictly within localized subgraphs, effectively preserving local heterogeneity and mitigating the oversmoothing problem. Extensive empirical evaluations conducted on six real-world networks demonstrate the superiority of TSLSG and validate the effectiveness of its individual components.
The findings of the study indicate that a large number of women in Bangladesh are suffering from being overweight or obese, and multiple factors are responsible for this including, older age,being from wealthy households, higher education, being from food-secured households, watching TV at least once a week, and being...
Haribondhu Sarma, N. Saquib, M. Hasan et al.· BMC Obesity· 75 citations· ⚡5
The rapid growth of scientific literature makes it increasingly challenging for researchers to identify novel and impactful ideas, especially across disciplines. Modern artificial intelligence (AI) systems offer new opportunities for scientific ideation, but how compelling are AI-generated ideas, and how can their qual...
Xuemei Gu, Mario Krenn· Proceedings of the National...· 65 citations
O Ongoing pro-poor health-financing strategies should be strengthened with introduction of innovative ways to monitor access, equity, and quality of care in maternal health.
Herfina Y. Nababan, M. Hasan, Tiara Marthias et al.· International Journal of Wom...· 64 citations· ⚡2
Agile methods continue to gain popularity. In particular, the Scrum method appears to be on the verge of becoming a de-facto standard in the industry, leading the so called Agile movement. While there are success stories and recommendations, there is little scientifically valid evidence of the challenges in the adoptio...
A. Marchenko, P. Abrahamsson· Agile Conference· 59 citations· ⚡11
Findings demonstrate a strong positive association between mother’s education and IYCF indicators and suggest strengthening national policies to educate women at least to the secondary level in SACs might be a cost-effective intervention for improving IY CF practices.
M. Tariqujjaman, M. Hasan, M. Mahfuz et al.· Nutrients· 45 citations· ⚡4
A comprehensive taxonomy of the challenges faced when a medium-scale organization decided to adopt software platforms is provided, namely: business challenges, organizational challenges, technical challenges, and people challenges.
Yaser Ghanam, F. Maurer, P. Abrahamsson· Information and Software Tec...· 41 citations· ⚡3
Machine-learning networks can learn without detailed instructions from a designer. Those that do may offer insight into how the brain reorganizes itself during learning.
MIT News · Artificial Intelligence· news.mit.eduJul 15, 2026
Assistant Professor Pat Pataranutaporn describes a new interface that lets everyday users glimpse inside an AI's neural network before their chatbot ever says a word.
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.