PRESOL: a web-based computational setting for feature-based flare forecasting

2025-10-02

AI Summary

This study addresses the limited interpretability and operational usability of solar flare prediction models by developing a web-based, feature-driven machine learning computing platform. The platform integrates automated feature engineering, multi-model training (including XGBoost and Random Forest), SHAP-based interpretability analysis, and comprehensive performance evaluation. It enables end-to-end flare probability prediction, key feature ranking, and interactive model comparison visualization upon user upload of solar active-region physical parameters. Its key innovation lies in the first online, open-access integration of feature selection with explainable AI (XAI), significantly enhancing transparency, reproducibility, and practicality in space weather forecasting. Experimental results demonstrate robust performance across multiple datasets, achieving AUC > 0.92, while consistently identifying dominant physical features—including magnetic flux and shear angle—thereby demonstrating strong potential for operational deployment in real-world space weather services.

Technology Category

• Machine Learning: Feature Construction/Reformulation
• Humans and AI: Explainable AI (XAI) for Human Understanding
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web search models and ranking

Abstract

Solar flares are the most explosive phenomena in the solar system and the main trigger of the events' chain that starts from Coronal Mass Ejections and leads to geomagnetic storms with possible impacts on the infrastructures at Earth. Data-driven solar flare forecasting relies on either deep learning approaches, which are operationally promising but with a low explainability degree, or machine learning algorithms, which can provide information on the physical descriptors that mostly impact the prediction. This paper describes a web-based technological platform for the execution of a computational pipeline of feature-based machine learning methods that provide predictions of the flare occurrence, feature ranking information, and assessment of the prediction performances.

Problem

• Developing a web-based platform for solar flare forecasting using machine learning
• Providing feature ranking to enhance explainability of flare predictions
• Delivering flare occurrence predictions with performance assessment capabilities

Innovation

• Web-based platform for solar flare prediction
• Feature-based machine learning pipeline
• Provides prediction and feature ranking