Soukaina Filali Boubrahimi

Soukaina Filali Boubrahimi serves as an Assistant Professor in the Computer Science Department within the College of Engineering at Utah State University. Her academic appointment is based in the SER 332 building located at 4205 Old Main Hill, Logan, UT 84322-0001. She maintains a research-active position with a focus on computational methods for complex temporal data analysis.

Dr. Filali Boubrahimi's research program centers on time series analysis, machine learning, and space weather prediction, with particular emphasis on solar flare forecasting and counterfactual explanation systems. Her work bridges theoretical machine learning advancements with practical applications in heliophysics, hydrology, and social media analysis. The research portfolio demonstrates significant expertise in handling imbalanced temporal datasets, developing novel data augmentation techniques, and creating interpretable AI systems for critical prediction tasks.

Analysis of her recent publication trajectory reveals consistent contributions to counterfactual explanation frameworks for time series data (Info-CELS, M-cels, ACTS), space weather prediction systems (solar flare and energetic particle event forecasting), and generative modeling approaches (AVATAR, ChronoGAN). Her work frequently addresses the challenges of severely imbalanced datasets through contrastive learning and sophisticated preprocessing techniques, demonstrating methodological innovation in handling rare but critical space weather events.

While no specific awards are documented in the available information, her research program appears substantial based on the volume and quality of recent publications spanning multiple high-impact domains. The research demonstrates strong interdisciplinary connections between computer science, space physics, and environmental science.

Her laboratory activities focus on developing machine learning frameworks for temporal data analysis, with particular attention to space weather prediction systems. The research group appears to specialize in creating robust models for rare event prediction, explainable AI systems for time series classification, and novel data augmentation techniques for imbalanced temporal datasets. Current projects likely include the development of multimodal fusion approaches for solar energetic particle prediction and spatio-temporal modeling for hydrological applications.