PREDICTING PARKING BEHAVIOR OVER TIME WITH PERIODIC WEATHER-AWARE LSTM AND EVENT MECHANISM

Abstract: Since there are so many variables to take into account, like the weather, the time of day, the day of the week, and nearby events, it can be challenging to predict parking behaviour. Conventional approaches to predicting parking behaviour have limitations since they don't take into consideration the periodic nature of the data or the impact of events and the weather. In this paper, we suggest an LSTM (Long Short-Term Memory) for forecasting parking behaviour that includes an event mechanism and periodic weather awareness. To accurately capture the periodicity and event dependencies in parking data, the recommended solution combines LSTM and event approaches. To increase the prediction's accuracy, weather data is also added.

Keywords: In this paper, we suggest an LSTM for forecasting parking behaviour that includes an event mechanism and periodic weather awareness.