English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

We present a case study of solar flare forecasting by means of metadatafeature time series, by treating it as a prominent class-imbalance andtemporally coherent problem. Taking full advantage of pre-flare time series insolar active regions is made possible via the Space Weather Analytics for SolarFlares (SWAN-SF) benchmark dataset; a partitioned collection of multivariatetime series of active region properties comprising 4075 regions and spanningover 9 years of the Solar Dynamics Observatory (SDO) period of operations. Weshowcase the general concept of temporal coherence triggered by the demand ofcontinuity in time series forecasting and show that lack of properunderstanding of this effect may spuriously enhance models' performance. Wefurther address another well-known challenge in rare event prediction, namely,the class-imbalance issue. The SWAN-SF is an appropriate dataset for this, witha 60:1 imbalance ratio for GOES M- and X-class flares and a 800:1 for X-classflares against flare-quiet instances. We revisit the main remedies for thesechallenges and present several experiments to illustrate the exact impact thateach of these remedies may have on performance. Moreover, we acknowledge thatsome basic data manipulation tasks such as data normalization and crossvalidation may also impact the performance -- we discuss these problems aswell. In this framework we also review the primary advantages and disadvantagesof using true skill statistic and Heidke skill score, as two widely usedperformance verification metrics for the flare forecasting task. In conclusion,we show and advocate for the benefits of time series vs. point-in-timeforecasting, provided that the above challenges are measurably andquantitatively addressed.

How to Train Your Flare Prediction Model: Revisiting Robust Sampling of Rare Events