머신러닝을 활용한 서울시 스포츠 시설 매출액 예측 모델 개발 = Development of Sales Prediction Model for Sports Facilities in Seoul Using Machine Learning

다국어 초록 (Multilingual Abstract)

The purpose of this research is to grasp the sales predict of sports facilities in Seoul and the main factors related to the sales forecast by utilizing machine learning. To conduct the study collected daily sales data of sports facilities in Seoul from January 2019 to March 2023. Collected weather information on the day sales were made, whether it was a weekend, the season, and daily COVID-19 confirmed cases in Seoul, and input them as explanatory variables. machine learning model was used to cross-validate the predictive power. As a result of cross-validation, the XGboost model showed the highest predictive power, showing an accuracy of 68.4% when prediction the facility industry in the entire city of Seoul. An analysis of all facilities businesses showed that the weather factor was the most frequently used factor. It is judged that it can be useful for facility management.