Multivariate Time Series Data Set for Space Weather Data Analytics

• Activity Number: 81 - Statistical and Machine Learning Efforts on Solar Flare Predictions I
• Type: Topic-Contributed
• Date/Time: Monday, August 9, 2021 : 10:00 AM to 11:50 AM
• Sponsor: Section on Physical and Engineering Sciences
• Author(s): Rafal Angryk*
• Companies: Georgia State University

Abstract

Reliable and clean big data sets lie at the heart of statistical and machine learning efforts dedicated to the prediction of solar events. In this talk, we will present our freely available and cross-checked multivariate time series dataset extracted from solar photospheric vector magnetograms in Spaceweather HMI Active Region Patch series. Our SWAN-SF data set (https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/EBCFKM) contains over 4,000 data collections from active regions, observed between May 2010 and December 2018; it includes 51 flare-predictive parameters, and integrates over 10,000 flare reports. In this talk, we will discuss the methods we used for data collection, cleaning and pre-processing of the solar active region and flare data, and describe our data integration and sampling methodology.