# Predicting Solar Flares Using CNN and LSTM on Two Solar Cycles of Active Region Data

arXiv: https://arxiv.org/abs/2204.03710 · HTML: https://mlsanity.com/paper/2204.03710/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 66 references that are in the map.
- First posted on arXiv 7 Apr 2022.

## Builds on (most cited first, 5 of 5)

- [How to Train Your Flare Prediction Model: Revisiting Robust Sampling of Rare Events](https://mlsanity.com/paper/2103.07542/) (2021)
- [Predicting Solar Flares Using a Long Short-Term Memory Network](https://mlsanity.com/paper/1905.07095/) (2019)
- [Flare Prediction Using Photospheric and Coronal Image Data](https://mlsanity.com/paper/1708.01323/) (2017)
- [A Framework for Designing and Evaluating Solar Flare Forecasting Systems](https://mlsanity.com/paper/2005.02493/) (2020)
- [Transfer Learning in Spatial-Temporal Forecasting of the Solar Magnetic Field](https://mlsanity.com/paper/1911.03193/) (2019)

## Built on by (most cited first, 15 of 15)

- [Explaining Full-disk Deep Learning Model for Solar Flare Prediction using Attribution Methods](https://mlsanity.com/paper/2307.15878/) (2023)
- [Minimum-Risk Recalibration of Classifiers](https://mlsanity.com/paper/2305.10886/) (2023)
- [Explainable Deep Learning-based Solar Flare Prediction with post hoc Attention for Operational Forecasting](https://mlsanity.com/paper/2308.02682/) (2023)
- [Exploring Deep Learning for Full-disk Solar Flare Prediction with Empirical Insights from Guided Grad-CAM Explanations](https://mlsanity.com/paper/2308.15712/) (2023)
- [Operational solar flare forecasting via video-based deep learning](https://mlsanity.com/paper/2209.05128/) (2022)
- [Solar Flare Index Prediction Using SDO/HMI Vector Magnetic Data Products with Statistical and Machine Learning Methods](https://mlsanity.com/paper/2209.13779/) (2022)
- [Deep Space Weather Model: Long-Range Solar Flare Prediction from Multi-Wavelength Images](https://mlsanity.com/paper/2508.07847/) (2025)
- [Incorporating Polar Field Data for Improved Solar Flare Prediction](https://mlsanity.com/paper/2212.01730/) (2022)
- [Solar Active Region Magnetogram Image Dataset for Studies of Space Weather](https://mlsanity.com/paper/2305.09492/) (2023)
- [Magnetogram-to-Magnetogram: Generative Forecasting of Solar Evolution](https://mlsanity.com/paper/2407.11659/) (2024)
- [Ensemble Learning for CME Arrival Time Prediction](https://mlsanity.com/paper/2305.00258/) (2023)
- [Probabilistic solar flare forecasting using historical magnetogram data](https://mlsanity.com/paper/2308.15410/) (2023)
- [Large Scale Evaluation of Deep Learning-based Explainable Solar Flare Forecasting Models with Attribution-based Proximity Analysis](https://mlsanity.com/paper/2411.18070/) (2024)
- [Solar Flare Prediction Using Long Short-term Memory (LSTM) and Decomposition-LSTM with Sliding Window Pattern Recognition](https://mlsanity.com/paper/2507.05313/) (2025)
- [SolARED: Solar Active Region Emergence Dataset for Machine Learning Aided Predictions](https://mlsanity.com/paper/2601.13145/) (2026)

## Abstract (arXiv)

We consider the flare prediction problem that distinguishes flare-imminent active regions that produce an M- or X-class flare in the future 24 hours, from quiet active regions that do not produce any flare within ± 24 hours. Using line-of-sight magnetograms and parameters of active regions in two data products covering Solar Cycle 23 and 24, we train and evaluate two deep learning algorithms -- CNN and LSTM -- and their stacking ensembles. The decisions of CNN are explained using visual attribution methods. We have the following three main findings. (1) LSTM trained on data from two solar cycles achieves significantly higher True Skill Scores (TSS) than that trained on data from a single solar cycle with a confidence level of at least 0.95. (2) On data from Solar Cycle 23, a stacking ensemble that combines predictions from LSTM and CNN using the TSS criterion achieves significantly higher TSS than the "select-best" strategy with a confidence level of at least 0.95. (3) A visual attribution method called Integrated Gradients is able to attribute the CNN's predictions of flares to the emerging magnetic flux in the active region. It also reveals a limitation of CNN as a flare prediction method using line-of-sight magnetograms: it treats the polarity artifact of line-of-sight magnetograms as positive evidence of flares.