作者: Aaron K. Baughman , Stefan A. G. Van Der Stockt , Michael Perlitz , Brian M. O'Connell
DOI:
关键词:
摘要: Approaches for optimizing network demand forecasting models and topology using hyperparameter selection are provided. An approach includes defining a pool of features that usable in predict resources, wherein the at least one historical feature event feature. The also generating, computer device, an optimal model subset selected from features. further predicting future on model. additionally allocating resources based predicted demand.