Graph Neural Networks: A Bibliometric Mapping of the Research Landscape and Applications DOI Creative Commons
Annielle Mendes Brito da Silva, Natiele Carla da Silva Ferreira, Luiza Amara Maciel Braga

et al.

Information, Journal Year: 2024, Volume and Issue: 15(10), P. 626 - 626

Published: Oct. 11, 2024

Graph neural networks (GNNs) are deep learning algorithms that process graph-structured data and suitable for applications such as social networks, physical models, financial markets, molecular predictions. Bibliometrics, a tool tracking research evolution, identifying milestones, assessing current research, can help identify emerging trends. This study aims to map GNN applications, directions, key contributors. An analysis of 40,741 GNN-related publications from the Web Science Core Collection reveals rising trend in publications, especially since 2018. Computer Science, Engineering, Telecommunications play significant roles with focus on learning, graph convolutional machine learning. China USA combined account 76.4% publications. Chinese universities concentrate feature extraction, task analysis, whereas American The also highlights importance Chemistry, Physics, Mathematics, Imaging & Photographic Technology, their respective knowledge communities. In conclusion, bibliometric provides an overview showing growing interest across various disciplines, highlighting potential GNNs solving complex problems need continued collaboration.

Language: Английский

Post-hoc vs ante-hoc explanations: xAI design guidelines for data scientists DOI Creative Commons
Carl Orge Retzlaff, Alessa Angerschmid, Anna Saranti

et al.

Cognitive Systems Research, Journal Year: 2024, Volume and Issue: 86, P. 101243 - 101243

Published: May 6, 2024

The growing field of explainable Artificial Intelligence (xAI) has given rise to a multitude techniques and methodologies, yet this expansion created gap between existing xAI approaches their practical application. This poses considerable obstacle for data scientists striving identify the optimal technique needs. To address problem, our study presents customized decision support framework aid in choosing suitable approach use-case. Drawing from literature survey insights interviews with five experienced scientists, we introduce tree based on trade-offs inherent various approaches, guiding selection six commonly used tools. Our work critically examines prevalent ante-hoc post-hoc methods, assessing applicability real-world contexts through expert interviews. aim is equip policymakers capacity select methods that not only demystify decision-making process, but also enrich user understanding interpretation, ultimately advancing application settings.

Language: Английский

Citations

41
Trustworthy multi-phase liver tumor segmentation via evidence-based uncertainty DOI
Chuanfei Hu, Tianyi Xia, Ying Cui

et al.

Engineering Applications of Artificial Intelligence, Journal Year: 2024, Volume and Issue: 133, P. 108289 - 108289

Published: March 22, 2024

Language: Английский

Citations

24
Transformer models in biomedicine DOI Creative Commons
Sumit Madan, Manuel Lentzen,

Johannes Brandt

et al.

BMC Medical Informatics and Decision Making, Journal Year: 2024, Volume and Issue: 24(1)

Published: July 29, 2024

Deep neural networks (DNN) have fundamentally revolutionized the artificial intelligence (AI) field. The transformer model is a type of DNN that was originally used for natural language processing tasks and has since gained more attention various kinds sequential data, including biological sequences structured electronic health records. Along with this development, transformer-based models such as BioBERT, MedBERT, MassGenie been trained deployed by researchers to answer scientific questions originating in biomedical domain. In paper, we review development application analyzing biomedical-related datasets textual protein sequences, medical structured-longitudinal images well graphs. Also, look at explainable AI strategies help comprehend predictions models. Finally, discuss limitations challenges current models, point out emerging novel research directions.

Language: Английский

Citations

12
From scientific theory to duality of predictive artificial intelligence models DOI Creative Commons
Jürgen Bajorath

Cell Reports Physical Science, Journal Year: 2025, Volume and Issue: unknown, P. 102516 - 102516

Published: April 1, 2025

Language: Английский

Citations

1
A Practical tutorial on Explainable AI Techniques DOI Open Access
Adrien Bennetot, Ivan Donadello, A. Haouari

et al.

ACM Computing Surveys, Journal Year: 2024, Volume and Issue: 57(2), P. 1 - 44

Published: June 12, 2024

The past years have been characterized by an upsurge in opaque automatic decision support systems, such as Deep Neural Networks (DNNs). Although DNNs great generalization and prediction abilities, it is difficult to obtain detailed explanations for their behavior. As Machine Learning models are increasingly being employed make important predictions critical domains, there a danger of creating using decisions that not justifiable or legitimate. Therefore, general agreement on the importance endowing with explainability. EXplainable Artificial Intelligence (XAI) techniques can serve verify certify model outputs enhance them desirable notions trustworthiness, accountability, transparency, fairness. This guide intended be go-to handbook anyone computer science background aiming intuitive insight from accompanied out-of-the-box. article aims rectify lack practical XAI applying techniques, particular, day-to-day models, datasets use-cases. In each chapter, reader will find description proposed method well one several examples use Python notebooks. These easily modified applied specific applications. We also explain what prerequisites technique, user learn about them, which tasks they aimed at.

Language: Английский

Citations

7
Collaborative weighting in federated graph neural networks for disease classification with the human-in-the-loop DOI Creative Commons

Christian Hausleitner,

Heimo Mueller,

Andreas Holzinger

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Sept. 19, 2024

Language: Английский

Citations

7
Graph Artificial Intelligence in Medicine DOI
Ruth Johnson, Michelle M. Li, Ayush Noori

et al.

Annual Review of Biomedical Data Science, Journal Year: 2024, Volume and Issue: 7(1), P. 345 - 368

Published: May 15, 2024

In clinical artificial intelligence (AI), graph representation learning, mainly through neural networks and transformer architectures, stands out for its capability to capture intricate relationships structures within datasets. With diverse data—from patient records imaging—graph AI models process data holistically by viewing modalities entities them as nodes interconnected their relationships. Graph facilitates model transfer across tasks, enabling generalize populations without additional parameters with minimal no retraining. However, the importance of human-centered design interpretability in decision-making cannot be overstated. Since information localized transformations defined on relational datasets, they offer both an opportunity a challenge elucidating rationale. Knowledge graphs can enhance aligning model-driven insights medical knowledge. Emerging integrate pretraining, facilitate interactive feedback loops, foster human–AI collaboration, paving way toward clinically meaningful predictions.

Language: Английский

Citations

5
Grand Challenges of Smart Technology for Older Adults DOI Creative Commons
Jia Zhou,

Gavriel Salvendy,

Walter R. Boot

et al.

International Journal of Human-Computer Interaction, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 43

Published: Feb. 26, 2025

Smart devices, systems, and services are transforming various aspects of daily life, offering new opportunities for the well-being older adults. This article explores grand challenges associated with evolving smart technology aged population. Based on collective effort 13 experts, their insights were categorized into six adults: DISUSE (underutilization technology), USE (user knowledge, goal complexity, risk-potential tradeoff), MISUSE (overreliance ABUSE (in appropriate application TIME (evolving relationship between adults DOMAIN (potential barriers use in health, home, work). Each challenge is further elaborated through its components or emerging issues, leading to implications elderly-friendly technology. Addressing these requires collaborative efforts ensure that effectively enhances active healthy aging.

Language: Английский

Citations

0
On the disagreement problem in Human-in-the-Loop federated machine learning DOI Creative Commons

Matthias J. M. Huelser,

Heimo Mueller,

Natalia Díaz-Rodríguez

et al.

Journal of Industrial Information Integration, Journal Year: 2025, Volume and Issue: unknown, P. 100827 - 100827

Published: March 1, 2025

Language: Английский

Citations

0
Agent-in-the-Loop to Distill Expert Knowledge into Artificial Intelligence Models: A Survey DOI

Jiayuan Gao,

Yingwei Zhang, Yiqiang Chen

et al.

Research Square (Research Square), Journal Year: 2025, Volume and Issue: unknown

Published: April 8, 2025

Abstract Large-scale neural networks have revolutionized many general knowledge areas (e.g., computer vision and language processing), but are still rarely applied in expert healthcare), due to data sparsity high annotation expenses. Human-in-the-loop machine learning (HIL-ML) incorporates domain into the modeling process, effectively addressing these challenges.Recently, some researchers started using large models substitute for certain tasks typically performed by humans. Although limitations areas, after being trained on trillions of examples, they demonstrated advanced capabilities reasoning, semantic understanding, grounding, planning. These can serve as proxies human, which introduces new opportunities challenges HIL-ML area.Based above, we summarize a more comprehensive framework, Agent-in-the-Loop Machine Learning (AIL-ML), where agent represents both humans models. AIL-ML efficiently collaborate human model construct vertical AI with lower costs.This paper presents first review recent advancements this area. First, provide formal definition discuss its related fields. Then, categorize methods based processing development, providing definitions each, present representative works detail each category. Third, highlight relative applications AIL-ML. Finally, current literature future research directions.

Language: Английский

Citations

0