Design and Implementation of Campus Bathroom Prediction System Based on Prophet Algorithm

Time series model is an important method for forecasting data series in the time dimension, which is widely used in many fields such as finance, economy, climate, etc. However, traditional time series forecasting methods are often too complicated and have limited effects. In order to effectively predict the flow of people in campus bathrooms and optimize public resources and management models, this paper develops a campus bathroom prediction system based on Facebook's open-source Prophet time series prediction model, and it's composed of growth trend model, seasonal trend model and holiday model. It can accurately fit the non-linear periodic trend and forecast the campus bathroom flow in a simpler and more flexible way, which greatly improves the availability and accuracy of the traditional model. In addition, this paper designs and elaborates on the system functions, database construction and interactive pages of campus bathroom prediction from the perspective of system development. Experiments show that the campus bathroom prediction method based on the Prophet algorithm has the advantages of simplicity, flexibility, high accuracy and good practicability, which can scientifically improve the utilization of bathroom equipment and optimize student experience.