Solar Flare Prediction Using Advanced Feature Extraction, Machine Learning, and Feature Selection

Ahmed, Omar; Qahwaji, Rami; Colak, Tufan; Higgins, Paul A. T.; Gallagher, Peter T.; Bloomfield, D. Shaun

doi:10.1007/s11207-011-9896-1

Cited by 168 publications

(159 citation statements)

References 33 publications

Supporting

Mentioning

146

Contrasting

Order By: Relevance

“…Considering classes C, M, and X, some methods consider "Positive" forecasts for classes greater than or equal to "C" [7], others consider "Positive" for forecasts greater than or equal to a class M solar flare [8][9][10][11], and others forecast individual probabilities for each class (C, M, X).…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

SeMiner: A Flexible Sequence Miner Method to Forecast Solar Time Series

et al. 2018

View full text Add to dashboard Cite

X-rays emitted by the Sun can damage electronic devices of spaceships, satellites, positioning systems and electricity distribution grids. Thus, the forecasting of solar X-rays is needed to warn organizations and mitigate undesirable effects. Traditional mining classification methods categorize observations into labels, and we aim to extend this approach to predict future X-ray levels. Therefore, we developed the "SeMiner" method, which allows the prediction of future events. "SeMiner" processes X-rays into sequences employing a new algorithm called "Series-to-Sequence" (SS). It employs a sliding window approach configured by a specialist. Then, the sequences are submitted to a classifier to generate a model that predicts X-ray levels. An optimized version of "SS" was also developed using parallelization techniques and Graphical Processing Units, in order to speed up the entire forecasting process. The obtained results indicate that "SeMiner" is well-suited to predict solar X-rays and solar flares within the defined time range. It reached more than 90% of accuracy for a 2-day forecast, and more than 80% of True Positive (TPR) and True Negative (TNR) rates predicting X-ray levels. It also reached an accuracy of 72.7%, with a TPR of 70.9% and TNR of 79.7% when predicting solar flares. Moreover, the optimized version of "SS" proved to be 4.36 faster than its initial version.

show abstract

Section: Related Workmentioning

confidence: 99%

“…We find forecasting methods using Support Vector Machines [9,12,13], Artificial Neural Networks [7,10,12,14], C4.5 decision trees [5,11,[15][16][17], Naive Bayes [12] and Bayesian Networks [15]. The majority of the works tries different classifiers, looking for the one that achieves the best results according to the adopted criteria.…”

Section: Related Workmentioning

confidence: 99%

SeMiner: A Flexible Sequence Miner Method to Forecast Solar Time Series

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Typically, a set of scalar properties is derived from line of sight (LOS) or vector magnetogram and analyzed in a supervised classification context to derive which combination of properties is predictive of increased flaring activity (Leka & Barnes 2004;Guo et al 2006;Barnes et al 2007;Georgoulis & Rust 2007;Schrijver 2007;Falconer et al 2008;Song et al 2009;Huang et al 2010;Yu et al 2010;Lee et al 2012;Ahmed et al 2013;Bobra & Couvidat 2015). Examples of scalar properties include: sunspot area, total unsigned magnetic flux, flux imbalance, neutral line length, maximum gradients along the neutral line, or other proxies for magnetic connectivity within ARs.…”

Section: Contextmentioning

confidence: 99%

Image patch analysis of sunspots and active regions

Moon

Delouille

et al. 2016

J. Space Weather Space Clim.

View full text Add to dashboard Cite

Context. Separating active regions that are quiet from potentially eruptive ones is a key issue in Space Weather applications. Traditional classification schemes such as Mount Wilson and McIntosh have been effective in relating an active region large scale magnetic configuration to its ability to produce eruptive events. However, their qualitative nature prevents systematic studies of an active region's evolution for example. Aims. We introduce a new clustering of active regions that is based on the local geometry observed in Line of Sight magnetogram and continuum images. Methods. We use a reduced-dimension representation of an active region that is obtained by factoring the corresponding data matrix comprised of local image patches. Two factorizations can be compared via the definition of appropriate metrics on the resulting factors. The distances obtained from these metrics are then used to cluster the active regions. Results. We find that these metrics result in natural clusterings of active regions. The clusterings are related to large scale descriptors of an active region such as its size, its local magnetic field distribution, and its complexity as measured by the Mount Wilson classification scheme. We also find that including data focused on the neutral line of an active region can result in an increased correspondence between our clustering results and other active region descriptors such as the Mount Wilson classifications and the R-value. Conclusions. Matrix factorization of image patches is a promising new way of characterizing active regions. We provide some recommendations for which metrics, matrix factorization techniques, and regions of interest to use to study active regions.

show abstract

“…d ASAP's team at University of Bradford carried out joint research work with Trinity College Dublin (Ahmed et al 2013) to develop novel machine-learning and feature-selection algorithms to combine two of the recent developments in solar physics and space weather, which are ASAP and the Solar Monitor Active Region Tracker (SMART) system (Higgins et al 2011). SMART extracts, characterizes, and tracks the evolution of active regions across the solar disk using line-of-sight magnetograms and a combination of image processing techniques.…”

Section: Operational Modeling For Nowcasting and Forecasting Productsmentioning

confidence: 99%

“…In this work data mining and spatiotemporal association algorithms were developed to associate MFs with flares in order to differentiate flaring and non-flaring MFs and enable the application of machine-learning algorithms for flares prediction. The prediction performance is assessed using standard forecast-verification measures and compared with the prediction measures of ASAP (Ahmed et al 2013). d ASAP's team at University of Bradford carried out joint research with Royal Observatory of Belgium (ROB), Glasgow University, and Trinity College Dublin (Verbeeck et al 2013).…”

Section: Operational Modeling For Nowcasting and Forecasting Productsmentioning

confidence: 99%

Progress in space weather modeling in an operational environment

Tsagouri

Belehaki

Bergeot

et al. 2013

J. Space Weather Space Clim.

Self Cite

View full text Add to dashboard Cite

This paper aims at providing an overview of latest advances in space weather modeling in an operational environment in Europe, including both the introduction of new models and improvements to existing codes and algorithms that address the broad range of space weather's prediction requirements from the Sun to the Earth. For each case, we consider the model's input data, the output parameters, products or services, its operational status, and whether it is supported by validation results, in order to build a solid basis for future developments. This work is the output of the Sub Group 1.3 ''Improvement of operational models'' of the European Cooperation in Science and Technology (COST) Action ES0803 ''Developing Space Weather Products and services in Europe'' and therefore this review focuses on the progress achieved by European research teams involved in the action.

show abstract

Solar Flare Prediction Using Advanced Feature Extraction, Machine Learning, and Feature Selection

Cited by 168 publications

References 33 publications

SeMiner: A Flexible Sequence Miner Method to Forecast Solar Time Series

SeMiner: A Flexible Sequence Miner Method to Forecast Solar Time Series

Image patch analysis of sunspots and active regions

Progress in space weather modeling in an operational environment

Contact Info

Product

Resources

About