A Deep Learning Approach for the Prediction of Retail Store SalesAbstract—The purpose of this research is to construct asales prediction model for retail stores using the deep learningapproach, which has gained significant attention in the rapidlydeveloping field of machine learning in recent years. Usingsuch a model for analysis, an approach to store managementcould be formulated. The present study uses three years’ worthof point-of-sale (POS) data from a retail store to construct asales prediction model that, given the sales of a particular day,predicts the changes in sales on the following day. As a result,a deep learning model that considers the L1 regularizationachieved a sale forecasting accuracy rate of 86%. The productsat the retail store have been finely categorized. Even if theattributes of the product categories are increased in numberfrom tens to thousands, the predictive accuracy did not fallby more than about 7%. In contrast, the accuracy decreasedby around 13% when the logistic regression model was used.These results indicate that deep learning is highly suitable forconstructing models that include multi-attribute variables. Thepresent research demonstrates that deep learning is effectivefor analyzing the POS data of retail stores.Keywords-Deep Learning; Marketing; POS data; Sales Pre-diction Model; Logistic Regression.