Browse Articles

This study uses interpretable machine learning models, lasso and ridge regression with Shapley analysis, to identify key sales drivers for Corporación Favorita, Ecuador’s largest grocery chain. The results show that macroeconomic factors, especially labor force size, have the greatest impact on sales, though geographic and seasonal variables like city altitude and holiday proximity also play important roles. These insights can help businesses focus on the most influential market conditions to enhance competitiveness and profitability.

This study examined behavioral and psychological predictors of burnout among high school and university students in Pakistan using survey data and machine-learning models. Shorter sleep and greater mental fatigue—especially fatigue—were associated with higher burnout risk, while a Random Forest model successfully identified students at risk of burnout.

This study investigates skin tone bias in artificial intelligence models used for dermatological disease classification and evaluates a CycleGAN-based data augmentation approach to improve diagnostic performance on darker skin types. We generated synthetic dark-skinned images to enhance dataset diversity and compared model performance before and after augmentation. The results demonstrate that augmentation with synthetic dermatological images can help reduce disparities in diagnostic performance across skin tones, highlighting a practical strategy for improving fairness in dermatology AI systems.

The authors studied the ability of deep learning models to predict droughts in the midwestern United States.

Based on the success of deep learning, recent works have attempted to develop a waste classification model using deep neural networks. This work presents federated learning (FL) for a solution, as it allows participants to aid in training the model using their own data. Results showed that with less clients, having a higher participation ratio resulted in less accuracy degradation by the data heterogeneity.

The authors investigate the ability of machine learning models to developing new drug-like molecules by learning desired chemical properties versus simply generating molecules that similar to those in the training set.

In this paper, we measured the privacy budgets and utilities of different differentially private mechanisms combined with different machine learning models that forecast traffic congestion at future timestamps. We expected the ANNs combined with the Staircase mechanism to perform the best with every value in the privacy budget range, especially with the medium high values of the privacy budget. In this study, we used the Autoregressive Integrated Moving Average (ARIMA) and neural network models to forecast and then added differentially private Laplacian, Gaussian, and Staircase noise to our datasets. We tested two real traffic congestion datasets, experimented with the different models, and examined their utility for different privacy budgets. We found that a favorable combination for this application was neural networks with the Staircase mechanism. Our findings identify the optimal models when dealing with tricky time series forecasting and can be used in non-traffic applications like disease tracking and population growth.

This study, uses SimulATe to model changes in microbial diversity and community structure upon administering a cocktail of antibiotics versus a single antibiotic. The authors hypothesized that antibiotic cocktails, particularly those combining broad-spectrum drugs like tetracyclines and trimethoprims, would cause a more significant reduction in gut microbial diversity compared to single-drug treatments. The findings confirmed a greater loss of microbial diversity with combinatorial treatments compared to single-drug treatments. While individual antibiotics dynamically reshaped the surviving species of the microbiome, antibiotic cocktails frequently cleared all species of the gut microbiome.

The authors test the effects of pH level on different kinds of dental cement to model the long-term effects of reflux-induced stomach acid exposure.

Authors emphasize the challenges of manual tumor segmentation and the potential of deep learning models to enhance accuracy by automatically analyzing MRI scans.