AN EVALUATION FRAMEWORK FOR PUBLICATIONS ON ARTIFICIAL NEURAL NETWORKS IN SALES FORECASTING

摘要: Although artificial neural networks (ANN) promise superior performance in forecasting theory, they are not yet an established method business practice. The vast degrees of freedom to parameterize ANNs have lead countless heuristic approaches simplify modeling, training, network selection and evaluation implemented with varying success. Consequently, a systematic is required order identify successful heuristics derive sound guidelines ANN modeling from publications. As each domain imposes different for classification or point prediction on specific datasets, literature review conducted, identifying 2538 publications within the but only 32 them applicable sales forecasting. identified evaluated through framework regarding their validity reliability experiment design documentation, promote publications, recommendations future experiments possibly gaps current research