Shape-based Feature Engineering for Solar Flare Prediction. (arXiv:2012.14405v1 [astro-ph.SR])
<a href="http://arxiv.org/find/astro-ph/1/au:+Deshmukh_V/0/1/0/all/0/1">Varad Deshmukh</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Berger_T/0/1/0/all/0/1">Thomas Berger</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Meiss_J/0/1/0/all/0/1">James Meiss</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Bradley_E/0/1/0/all/0/1">Elizabeth Bradley</a>
Solar flares are caused by magnetic eruptions in active regions (ARs) on the
surface of the sun. These events can have significant impacts on human
activity, many of which can be mitigated with enough advance warning from good
forecasts. To date, machine learning-based flare-prediction methods have
employed physics-based attributes of the AR images as features; more recently,
there has been some work that uses features deduced automatically by deep
learning methods (such as convolutional neural networks). We describe a suite
of novel shape-based features extracted from magnetogram images of the Sun
using the tools of computational topology and computational geometry. We
evaluate these features in the context of a multi-layer perceptron (MLP) neural
network and compare their performance against the traditional physics-based
attributes. We show that these abstract shape-based features outperform the
features chosen by the human experts, and that a combination of the two feature
sets improves the forecasting capability even further.
Solar flares are caused by magnetic eruptions in active regions (ARs) on the
surface of the sun. These events can have significant impacts on human
activity, many of which can be mitigated with enough advance warning from good
forecasts. To date, machine learning-based flare-prediction methods have
employed physics-based attributes of the AR images as features; more recently,
there has been some work that uses features deduced automatically by deep
learning methods (such as convolutional neural networks). We describe a suite
of novel shape-based features extracted from magnetogram images of the Sun
using the tools of computational topology and computational geometry. We
evaluate these features in the context of a multi-layer perceptron (MLP) neural
network and compare their performance against the traditional physics-based
attributes. We show that these abstract shape-based features outperform the
features chosen by the human experts, and that a combination of the two feature
sets improves the forecasting capability even further.
http://arxiv.org/icons/sfx.gif
