Прогнозування подій в інтернет-торгівлі на основі ансамблевих моделей
Journal Title: Технічні вісті - Year 2018, Vol 1, Issue 47
Abstract
The application of machine learning methods for sales forecasting in online trading has been considered. The advantages and disadvantages of forecasting models on the basis of ensemble training have been shown. The results of application of considered methods to predict when items sell out have been presented.
Authors and Affiliations
Чан Чі Кионг, Ю. В. Цимбал
Keywords
Related Articles
- EP ID EP547359
- DOI -
- Views 96
- Downloads 0
Класифікація зображень на основі застосування автоасоціативних нейронних мереж
The problem of images classification is considered. It is proposed to use the autoassociative neural networks based on the geometric transformation model for this problem solving. The advantages and disadvantages of this...
Зміцнення відновленого поверхневого шару сталі 45 шляхом лазерного борування
By means of the proposed technologies, the relationship and strengthening of the optimal modes of recovery of the restoring layer and the optimal laser treatment modes with bat-retaining straps are chosen. The microstruc...
Алгоритм аналізу медичних послуг етапу “профілактика” за допомогою індикаторів якості
Formed an expanded set of indicators that characterize the quality of of medical services stage "prevention". For the optimization phase risks "prevention" of medical services is taken into account as a factor in the rel...
Особливості проектування силових трифазних трансформаторів
The article deals with the principles of designing power three-phase transformers. The correct calculation will allow to choose construction materials, will provide the optimal size of the transformer and maximum efficie...
Розробка системи прогнозування результатів спортивних подій на основі нейронних мереж
This product is a powerful solution aimed at obtaining a prediction of the outcome of a football match. It also includes a database of past match statistics for more realistic and accurate prediction. The introduction of...