Tag: Explainable AI

A simulated 10-year study found that AI forecasting and flexible investment timing raised modeled wastewater biorefinery value from negative $1.2 ...

A hierarchical deep-learning framework classified five lung and colon tissue classes with 98.50% accuracy on LC25000 and 82.01% on the ...

A systematic review of 168 primary studies found that 57.14% did not examine the performance–faithfulness relationship, 41.07% reported synergy and ...

A machine-learning study using 2011–2025 data from Zambia’s regulated fuel-pricing system found that Ridge regression outperformed Random Forest, Support Vector ...

A new leakage-controlled machine learning framework achieves strong sentiment classification on social media data while revealing that engagement prediction collapses ...

A cross-validation study of 1,090 images from one public orange dataset found 99.12% ± 0.71% accuracy for ViT + PCA ...

A leakage-aware evaluation of explainable machine learning found modest prediction of nanoparticle tumor delivery, with errors increasing for unseen materials ...

A divergence-aware federated framework achieved 95.75% accuracy on non-IID email-malware benchmarks while generating visual explanations for its predictions.

Researchers in China built an explainable XGBoost model using six routine admission variables to predict which mild stroke patients will ...

A lean XAI framework combining SHAP, SHAPTime, and rolling-origin validation delivers stable, horizon-aware explanations for supply-chain demand forecasts on the ...

Researchers used SHAP-based explainable AI to identify a single EEG electrode and just seven features that detect virtual reality motion ...

Researchers combined deep learning, grey wolf optimization, explainable heatmaps and a Gemini-powered chatbot to detect finger millet diseases with 98.36 ...

Researchers at Amrita Vishwa Vidyapeetham built an explainable machine learning framework that predicts student dropout risk from academic, financial, health, ...

A new explainable transformer that combines frequency-domain analysis and topology-aware structural modeling achieves 97.86 percent accuracy in classifying eight retinal ...

Researchers in Naples have developed EDOKS, a transparent image similarity metric that combines Earth Mover's Distance texture analysis with Oklab ...

Malaysian researchers have built a hybrid CNN-LSTM traffic forecasting system that pairs deep learning accuracy with Formal Concept Analysis to ...

A unified framework combining federated learning, differential privacy and explainable AI tackles drift, fairness and privacy in credit scoring, revealing ...

A sweeping review finds AI delivers impressive agricultural gains in irrigation, soil management, and pest detection, but warns that spatial ...

A transformer-based deep-learning pipeline called ProFormer classifies single-cell and plasma proteomes directly from raw mass-spectrometry fingerprints, outperforming CNNs and enabling ...

Deep learning models that promise to weigh pastures from photographs may be far less trustworthy than their published accuracy figures ...