Bibliographic citations
Valencia, J., Flores, D. (2023). Modelo híbrido para la toma de decisiones y predicción de ingresos en empresas del rubro de tragamonedas aplicando XGBoost” [Tesis, Universidad Peruana de Ciencias Aplicadas (UPC)]. http://hdl.handle.net/10757/669582
Valencia, J., Flores, D. Modelo híbrido para la toma de decisiones y predicción de ingresos en empresas del rubro de tragamonedas aplicando XGBoost” [Tesis]. PE: Universidad Peruana de Ciencias Aplicadas (UPC); 2023. http://hdl.handle.net/10757/669582
@misc{renati/408582,
title = "Modelo híbrido para la toma de decisiones y predicción de ingresos en empresas del rubro de tragamonedas aplicando XGBoost”",
author = "Flores Osorio, Dalia Denisse",
publisher = "Universidad Peruana de Ciencias Aplicadas (UPC)",
year = "2023"
}
Due to the recent COVID 19 pandemic, the casino and slot machine sector has been forced to incorporate new techniques for information analysis. Currently, there are companies that continue to process information and obtain reports manually, which makes strategic and operational decision-making difficult on a large scale. Different investigations have been carried out on decision-making in the sales process, but these have not focused on the casino business. A hybrid model is proposed for decision making and income prediction in companies in the slots industry applying XGBoost. The proposal was developed applying the Ralph Kimball methodology, made up of 6 phases: Project Planning, Definition of Business Requirements, Implementation of ETL processes, Design of Dimensional Modeling, Design of the Power BI Model, Prediction and forecast. The proposed model was validated through a case study of a casino company in Lima, Peru. Dashboards were created in Power BI to show revenue by company and brand. To obtain the precision, the comparison of 3 Machine Learning algorithms was made: XGBoost, Linear Regression and Support Vector Regression. The results showed that the XGBoost algorithm obtained better performance for the prediction and forecast of income, obtaining an accuracy of 0.93.
This item is licensed under a Creative Commons License