AI Meets Radar Echoes to Pinpoint Wiring Faults With Near-Perfect Accuracy
A French research team has combined time-domain reflectometry with temporal convolutional networks to detect, locate, and characterize wiring faults, achieving ...
Tag: Machine Learning
DistilBERT Outperforms Perceiver in the Hunt for Fake Hotel Reviews
A comparative study finds that a fine-tuned DistilBERT pipeline reliably detects deceptive hotel reviews while a Perceiver-based pipeline collapses toward ...
Your Phone Knows When You Are Stressed: KAIST Turns Everyday App Habits Into a Mental Health Signal
KAIST researchers developed a technology called SHIRBT-MMF that identifies daily stress from repeated smartphone behaviors such as messenger chatting, web ...
Wealth, Not Diet, Drives Global Thyroid Cancer Rates, Machine Learning Study Finds
A global machine learning study of 163 countries finds that national wealth and diagnostic intensity, not diet, explain most of ...
AI Meets the Mind: How Health Data Science Is Rewiring Neuroscience and Psychology
Artificial intelligence is quietly transforming two of the most intimate fields of medicine: neuroscience and psychology. A new editorial published ...
How a Hidden Flaw in EEG Machine Learning Inflates Accuracy by 35 Percentage Points
A new study shows that randomly splitting EEG windows between training and test sets inflates diagnostic accuracy by roughly 35 ...
AI Model Flags Septic Shock Risk in Appendicitis Patients
A multicenter machine learning study of 3,000 appendicitis patients shows that routine blood indices and ultrasound signs of barrier failure ...
Why Arabic Text-to-Speech Resists Easy Fixes: Inside a Pronunciation-Sensitive Stress Test of XTTS
A controlled adaptation campaign for Modern Standard Arabic speech synthesis finds that synthetic diacritization, punctuation-guided length rescue, and longer training ...
Extra-P Turns Scalability Modeling Into a Few Small Experiments
A decade-long effort at ETH Zurich and TU Darmstadt has distilled empirical performance modeling into a Python tool that finds ...
Adaptive controlled mutual learning framework for text-image person re-identification
Adaptive controlled mutual learning framework for text-image person re-identification
Blood Pressure Drug Losartan Awakens Metabolic Brown Fat in Surprise Discovery
Rockefeller University researchers found that the common blood pressure drug losartan stimulates metabolically active brown fat in mice and is ...
AI and Multi-Omics Decode the Rhizosphere Microbiome Against Southern Blight Fungus
A systematic review synthesizes how metagenomics, metatranscriptomics, metabolomics, proteomics, and machine learning converge to reveal and engineer rhizosphere microbial communities ...
AI Flags Hidden Self-Harm History in Europe’s Emergency Rooms
An interpretable machine learning model trained on European injury surveillance data can identify emergency department patients whose records show documented ...
Machine Learning Reads Facial Cues to Predict Speed Dating Decisions
Researchers trained machine learning models on minute-by-minute facial, audio, and linguistic features from real speed dating conversations, finding that facial ...
3D-printed high-entropy alloys are rewriting the rules of metal strength
A new review in Frontiers in Materials details how 3D-printing technologies are unlocking the extraordinary potential of high-entropy alloys, from ...
Neural Networks Learn to Chart Spacecraft Trajectories Faster Than Classic Math
Researchers at KTH Royal Institute of Technology show that geometry-aware normalization lets simple neural networks rapidly approximate Lambert transfer solutions ...
Machine Learning Predicts EV Battery Temperature on Cells It Has Never Seen
Researchers in India have built a machine-learning framework that predicts lithium-ion battery temperature during discharge with around one degree Celsius ...
Federated learning brings privacy-first anomaly detection to smart grids
Smart grids promise a cleaner, more responsive electricity system, but they also carry a hidden vulnerability: every smart meter is ...
DepressNet: An Interpretable Machine Learning Framework for Stratifying Depression Severity From Psychometric Data
Researchers have built DepressNet, an interpretable stacking-ensemble machine learning framework that stratifies depression severity into five levels from DASS-42 questionnaire ...
AI Forensic Accounting Screens Palestinian Banks for Hidden Instability Risk
Machine learning and SHAP explainability reveal conditional disclosure-stability patterns across seven Palestinian banks, offering supervisors an early-warning triage tool.
