MODEL FORECASTING TREN KUNJUNGAN WISATAWAN DI DIY MENGGUNAKAN REGRESI LOGISTIK BINER

The Special Region of Yogyakarta (DIY) is a province that has special features in terms of arts and culture, nature, culinary, shopping, and other tourism. Visits by both foreign and domestic tourists to DIY before the COVID pandemic and after the pandemic have fluctuated. The many choices of tourist attractions in DIY attract tourists to come. An increase or decrease in the number of tourists can affect various aspects of people's lives. An increase in tourist visits will affect congestion and traffic on the highway, while a decrease in the number of visits can affect regional income and the income of people who depend on tourism for their livelihood. In addition, the increase in the number of tourists also needs to be anticipated by the government to provide road infrastructure, buildings, city planning, traffic order, waste management, and so on. The existence of a forecasting model that can predict future tourist visits can help stakeholders make decisions to handle problems related to the impact of tourism on the community and to improve the governance of tourist visits. The use of a binary logistic regression algorithm in this case is used to predict the trend of an increasing or decreasing number of tourists for the next two years until 2026. Historical data of tourist visits from 2018 to 2024 from BPS is used for this study. The forecasting results show an increase in tourist visits in June 2025 and 2026. The evaluation results show the accuracy, precision, recall, and f1-score values of 1.0 (for the range 0-1). These results indicate that the forecasting model has a very good accuracy value.