# Solar flare forecasting from magnetic feature properties generated by Solar Monitor Active Region Tracker

## TL;DR

This study evaluates magnetic feature properties from SMART data for solar flare prediction, demonstrating that simple linear models with selected features outperform previous results and complex models are unnecessary.

## Contribution

It introduces the use of marginal relevance for feature selection and applies logistic regression for flare forecasting, achieving superior results over prior studies.

## Key findings

- Linear model with three features achieves TSS=0.84
- Competitive TSS=0.67 with NOAA AR subset
- Complex models do not improve prediction performance

## Abstract

We study the predictive capabilities of magnetic feature properties (MF) generated by Solar Monitor Active Region Tracker (SMART) for solar flare forecasting from two datasets: the full dataset of SMART detections from 1996 to 2010 that has been previously studied by Ahmed et al. (2011) and a subset of that dataset which only includes detections that are NOAA active regions (ARs). Main contributions: we use marginal relevance as a filter feature selection method to identify most useful SMART MF properties for separating flaring from non-flaring detections and logistic regression to derive classification rules to predict future observations. For comparison, we employ a Random Forest, Support Vector Machine and a set of Deep Neural Network models, as well as Lasso for feature selection. Using the linear model with three features we obtain significantly better results (TSS=0.84) to those reported by Ahmed et al.(2011) for the full dataset of SMART detections. The same model produced competitive results (TSS=0.67) for the dataset of SMART detections that are NOAA ARs which can be compared to a broader section of flare forecasting literature. We show that more complex models are not required for this data.