Details

Autor(en) / Beteiligte

• Pozdnoukhov, A.
• Purves, R.S.
• Kanevski, M.

Titel

Applying machine learning methods to avalanche forecasting

Ist Teil von

• Annals of glaciology, 2008, Vol.49, p.107-113

Ort / Verlag

Cambridge, UK: Cambridge University Press

Erscheinungsjahr

2008

Quelle

EZB-FREE-00999 freely available EZB journals

Beschreibungen/Notizen

• Avalanche forecasting is a complex process involving the assimilation of multiple data sources to make predictions over varying spatial and temporal resolutions. Numerically assisted forecasting often uses nearest-neighbour methods (NN), which are known to have limitations when dealing with high-dimensional data. We apply support vector machines (SVMs) to a dataset from Lochaber, Scotland, UK, to assess their applicability in avalanche forecasting. SVMs belong to a family of theoretically based techniques from machine learning and are designed to deal with high-dimensional data. Initial experiments showed that SVMs gave results that were comparable with NN for categorical and probabilistic forecasts. Experiments utilizing the ability of SVMs to deal with high dimensionality in producing a spatial forecast show promise, but require further work.