Show simple item record

dc.contributor.authorLotz, Stefan
dc.contributor.authorBeukes, Jacques P.
dc.contributor.authorDavel, Marelie H.
dc.date.accessioned2020-01-27T13:22:32Z
dc.date.available2020-01-27T13:22:32Z
dc.date.issued2019-12
dc.identifier.citationStefan Lotz, Jacques P. Beukes and Marelie H. Davel, “Input parameter ranking for neural networks in a space weather regression problem“, In Proc. South African Forum for Artificial Intelligence Research (FAIR2019), pp133-144, Cape Town, South Africa, December 2019.en_US
dc.identifier.issn1613-0073
dc.identifier.urihttp://hdl.handle.net/10394/33956
dc.description.abstractGeomagnetic storms are multi-day events characterised by significant perturbations to the magnetic field of the Earth, driven by so-lar activity. Numerous efforts have been undertaken to utilise in-situ mea-surements of the solar wind plasma to predict perturbations to the geo-magnetic field measured on the ground. Typically, solar wind measure-ments are used as input parameters to a regression problem tasked with predicting a perturbation index such as the 1-minute cadence symmetric-H (Sym-H) index. We re-visit this problem, with two important twists: (i) An adapted feedforward neural network topology is designed to en-able the pairwise analysis of input parameter weights. This enables the ranking of input parameters in terms of importance to output accu-racy, without the need to train numerous models. (ii) Geomagnetic storm phase information is incorporated as model inputs and shown to increase performance. This is motivated by the fact that different physical phe-nomena are at play during different phases of a geomagnetic storm.en_US
dc.language.isoenen_US
dc.publisherIn Proc. South African Forum for Artificial Intelligence Research (FAIR2019)en_US
dc.subjectSpace weatheren_US
dc.subjectInput parameter selectionen_US
dc.subjectNeural networken_US
dc.titleInput parameter ranking for neural networks in a space weather regression problemen_US
dc.typeOtheren_US


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record