Transformer-Based Named Entity Recognition in Construction Supply Chain Risk Management in Australia

IEEE Access, Год журнала: 2024, Номер 12, С. 41829 - 41851

Опубликована: Янв. 1, 2024

In the Australian construction industry, effective supply chain risk management (SCRM) is critical due to its complex networks and susceptibility various risks. This study explores application of transformer models like BERT, RoBERTa, DistilBERT, ALBERT, ELECTRA for Named Entity Recognition (NER) in this context. Utilizing these models, we analyzed news articles identify classify entities related risks, providing insights into vulnerabilities within sector. Among evaluated RoBERTa achieved highest average F1 score 0.8580, demonstrating superior balance precision recall NER Our findings highlight potential NLP-driven solutions revolutionize SCRM, particularly geo-specific settings.

Язык: Английский

---

Human attention-guided explainable artificial intelligence for computer vision models

Neural Networks, Год журнала: 2024, Номер 177, С. 106392 - 106392

Опубликована: Май 15, 2024

Explainable artificial intelligence (XAI) has been increasingly investigated to enhance the transparency of black-box models, promoting better user understanding and trust. Developing an XAI that is faithful models plausible users both a necessity challenge. This work examines whether embedding human attention knowledge into saliency-based methods for computer vision could their plausibility faithfulness. Two novel object detection namely FullGrad-CAM FullGrad-CAM++, were first developed generate object-specific explanations by extending current gradient-based image classification models. Using as objective measure, these achieve higher explanation plausibility. Interestingly, all when applied generally produce saliency maps are less model than from same task. Accordingly, attention-guided (HAG-XAI) was proposed learn how best combine explanatory information using trainable activation functions smoothing kernels maximize similarity between map map. The evaluated on widely used BDD-100K, MS-COCO, ImageNet datasets compared with typical perturbation-based methods. Results suggest HAG-XAI enhanced trust at expense faithfulness it plausibility, faithfulness, simultaneously outperformed existing state-of-the-art

Язык: Английский

---

Explainable artificial intelligence approaches for brain–computer interfaces: a review and design space

Journal of Neural Engineering, Год журнала: 2024, Номер 21(4), С. 041003 - 041003

Опубликована: Июль 20, 2024

This review paper provides an integrated perspective of Explainable Artificial Intelligence (XAI) techniques applied to Brain-Computer Interfaces (BCIs). BCIs use predictive models interpret brain signals for various high-stake applications. However, achieving explainability in these complex is challenging as it compromises accuracy. Trust can be established by incorporating reasoning or causal relationships from domain experts. The field XAI has emerged address the need across stakeholders, but there a lack BCI (XAI4BCI) literature. It necessary differentiate key concepts like explainability, interpretability, and understanding, often used interchangeably this context, formulate comprehensive framework.

Язык: Английский

---

Diagnostic Performance Evaluation of Multiparametric Magnetic Resonance Imaging in the Detection of Prostate Cancer with Supervised Machine Learning Methods

Diagnostics, Год журнала: 2023, Номер 13(4), С. 806 - 806

Опубликована: Фев. 20, 2023

Prostate cancer is the second leading cause of cancer-related death in men. Its early and correct diagnosis particular importance to controlling preventing disease from spreading other tissues. Artificial intelligence machine learning have effectively detected graded several cancers, prostate cancer. The purpose this review show diagnostic performance (accuracy area under curve) supervised algorithms detecting using multiparametric MRI. A comparison was made between performances different machine-learning methods. This study performed on recent literature sourced scientific citation websites such as Google Scholar, PubMed, Scopus, Web Science up end January 2023. findings reveal that techniques good with high accuracy curve for prediction MR imaging. Among methods, deep learning, random forest, logistic regression appear best performance.

Язык: Английский

---

Interpretation and explanation of convolutional neural network-based fault diagnosis model at the feature-level for building energy systems

Energy and Buildings, Год журнала: 2023, Номер 295, С. 113326 - 113326

Опубликована: Июнь 27, 2023

Язык: Английский

---