Paper Info
Title
Understanding The Impact Of Statistical Time Series Features For Flare Prediction Analysis
Abstract
Machine learning-based space weather analytics has attracted much attention due to the potential damages that. can be caused by the extreme space weather events. Using a recently released data benchmark, named SWAN-SF, designed for solar flare forecasting based on the pre-flare time series of solar magnetic field parameters, we conduct a case study on the impacts of statistical features derived from the multivariate time series. We investigate the relationship between the number of needed statistical features extracted from the multi-variate time series and the performance of flare forecast models. To that end, we employ random forest and mean decrease impurity to determine a feature selection methodology along with an evaluation procedure. The proposed evaluation method delivers a balance between the two frequently used metrics in this domain, namely True Skill Statistic and Heidke Skill Score. Our approach allows to introduce a generic feature selection and evaluation procedure that is independent from the minor and often obscured decisions that must he made for having a binary forecast model, while presenting interpretable and actionable tools that can help non-data experts make more informed and realistic decisions.
Year
DOI
Venue
2019
10.1109/BigData47090.2019.9006116
2019 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA)
Keywords
Field
DocType
time series classification, feature selection, statistical time series features, flare prediction
Data mining,Forecast skill,Feature selection,Statistic,Computer science,Multivariate statistics,Artificial intelligence,Analytics,Random forest,Machine learning,Space weather,Binary number
Conference
ISSN
Citations 
PageRank 
2639-1589
0
0.34
References 
Authors
0
6
Name
Order
Citations
PageRank
Maxwell Hostetter100.34
Azim Ahmadzadeh212.39
Berkay Aydin34010.75
Manolis K. Georgoulis401.01
dustin kempton5126.54
Rafal A. Angryk627145.56