DOBRO : a prediction error correcting robot under drifts

A. Maslov, H.T. Lam, M. Pechenizkiy, E. Bouillet, T. Kärkkäinen

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review


We propose DOBRO, a light online learning module, which is equipped with a smart correction policy helping making decision to correct or not the given prediction depending on how likely the correction will lead to a better prediction performance. DOBRO is a standalone module requiring nothing more than a time series of prediction errors and it is flexible to be integrated into any black-box model to improve its performance under drifts. We performed evaluation in a real-world application with bus arrival time prediction problem. The obtained results show that DOBRO improved prediction performance significantly meanwhile it did not hurt the accuracy when drift does not happen
Originele taal-2Engels
TitelSAC '16 Proceedings of the 31st Annual ACM Symposium on Applied Computing, 4-8 April 2016, Pisa, Italy
Plaats van productieNew York
UitgeverijAssociation for Computing Machinery, Inc
Aantal pagina's4
ISBN van geprinte versie978-1-4503-3739-7
StatusGepubliceerd - 2016

Vingerafdruk Duik in de onderzoeksthema's van 'DOBRO : a prediction error correcting robot under drifts'. Samen vormen ze een unieke vingerafdruk.

Citeer dit