This study mapped land use changes in the Grombalia Region of Tunisia using Sentinel-2 imagery and machine learning. Three classifiers—Random Forest, SVM, and CNN—were compared. Random Forest achieved the highest accuracy. Results highlight agricultural changes over time and demonstrate the effectiveness of remote sensing for environmental monitoring.

This study developed a real-time IoT-based system to optimize fishway performance in fragmented rivers. Using sensors, PIT-tag tracking, and machine-learning models, it links climate triggers with hydraulic controls. Adaptive sluice-gate regulation improved fish passage efficiency by 166% without reducing hydropower output, offering scalable, sustainable river management.

This research applies machine learning to genetic data to distinguish harmless DNA variations from cancer-causing mutations. By treating DNA like a crime scene, the model learns to identify which genetic changes truly drive breast cancer risk, supporting more accurate diagnosis and informed clinical decision-making.

Using machine learning and Hidden Markov Models, this research analyzes the authorship of disputed New Testament letters. The results show that stylistic differences reflect the Apostle Paul’s versatile writing styles rather than forgery, demonstrating how modern computational tools can help recover long-standing historical truths.

This research exposes a hidden privacy risk in online voting and video conferencing: eye movements captured by standard webcams can reveal user choices. Using AI models, voting decisions were inferred with over 95% accuracy, highlighting that digital security must address behavioral signals—not just encryption.

This research develops DNA-origami-enhanced nanopores to detect individual biomolecules from a single drop of blood. By slowing molecules and reading their electrical signatures with machine learning, the technology enables rapid, ultra-early disease diagnosis without traditional laboratory testing.

Inspired by biological reproduction, this research uses evolutionary algorithms to evolve mathematical equations that describe physical systems. Unlike black-box AI, these models are transparent and adaptable. By combining evolution with graph neural networks, the approach improves simulations for applications such as traffic control, robotics, and engineering design.

Partner choice increasingly reflects shared career aspirations, intensifying income inequality. Using Danish registry data and machine learning, this research shows assortative matching by education and career focus has risen since the 1980s. If pairing patterns had remained unchanged, today’s income inequality would be over 40% lower, highlighting family formation as a key economic force.

Lead contamination in drinking water threatens millions. This research combines physics-based pipe models with machine learning to identify lead pipes using vibration data. Generating thousands of simulated signals enabled a classifier with 99% accuracy, offering a noninvasive, cost-effective method to locate hidden lead pipes and support safer water infrastructure worldwide.

This research shows that genetic risk scores alone are insufficient for predicting chronic disease. By incorporating social and environmental factors using machine learning, disease prediction improves substantially, especially for disadvantaged populations. Integrating genetic and social risk is essential for equitable, effective personalized medicine.