Prediction of Solar Flare Size and TimetoFlare Using Support Vector Machine Regression
Abstract
We study the prediction of solar flare size and timetoflare using 38 features describing magnetic complexity of the photospheric magnetic field. This work uses support vector regression to formulate a mapping from the 38dimensional feature space to a continuousvalued label vector representing flare size or timetoflare. When we consider flaring regions only, we find an average error in estimating flare size of approximately half a geostationary operational environmental satellite (GOES) class. When we additionally consider nonflaring regions, we find an increased average error of approximately threefourths a GOES class. We also consider thresholding the regressed flare size for the experiment containing both flaring and nonflaring regions and find a true positive rate of 0.69 and a true negative rate of 0.86 for flare prediction. The results for both of these size regression experiments are consistent across a wide range of predictive time windows, indicating that the magnetic complexity features may be persistent in appearance long before flare activity. This is supported by our larger error rates of some 40 hr in the timetoflare regression problem. The 38 magnetic complexity features considered here appear to have discriminative potential for flare size, but their persistence in time makes them less discriminative for the timetoflare problem.
 Publication:

The Astrophysical Journal
 Pub Date:
 October 2015
 DOI:
 10.1088/0004637X/812/1/51
 arXiv:
 arXiv:1511.01941
 Bibcode:
 2015ApJ...812...51B
 Keywords:

 methods: data analysis;
 methods: statistical;
 Sun: flares;
 Sun: magnetic fields;
 Sun: photosphere;
 Astrophysics  Solar and Stellar Astrophysics
 EPrint:
 http://iopscience.iop.org/article/10.1088/0004637X/812/1/51/meta