The income prediction module of the retail store's network

Abstract

The main idea of this paper focused on the development of a program module, which predicts the pharmacy retail income by the machine learning theory. Beyond that, we want to introduce the best prediction model, which has learned by speci c retail dataset. Notice, the architecture of program involves dynamic upload dataset, by Yandex" Internet service. The dataset represents the set of features and set of retail points, however in this task, the features describe the pharmacy industry. So on the rst step will analyze the dataset and found out the correlation of the features. Next, will select the relevant features, which a ect on income rate of the retail point. The last one will introduce to the prediction income Average model. In the last, will compare the three models, there is Average model, Gradient Boosting Regression and Random Forest Regression