Parking lot occupancy prediction using long short-term memory and statistical methods

In crowded city centers, drivers looking for available parking space generate extra traffic and in addition, the resulting excessive exhaust gases cause air pollution. Therefore, directing to a spot an intelligent way is important task smart applications. This requires prediction of occupancy states lots which involves appropriate processing historical data. this work, Long-Short Term Memory (LSTM) Autoregressive Integrated Moving Average (ARIMA) methods were applied data collected from curbside spots Adana, Turkey predicting lot rates future values. The experiments performed making predictions with different horizons that are 1 minute, 5 minutes, 15 minutes. performances compared by calculating root mean squared error (RMSE) absolute (MAE) on five days. According results, when horizon set LSTM achieved RMSE MAE values 0.98 0.72, respectively. For same horizon, ARIMA 0.62 0.35, On other hand, smaller larger horizons. conclusion, it was shown more suitable horizons, however, better at near-future