<![CDATA[The increasing number of motor vehicles in urban areas has contributed to declining air quality, affecting both public health and the environment. This condition becomes more critical at road intersections with high traffic density, particularly during morning rush hours. This study aims to develop a real-time air quality prediction system based on Internet of Things and the Long Short-Term Memory (LSTM) method. Air quality data were collected using IoT-based sensors installed at a road intersection with a traffic density of approximately 200 motor vehicles per minute between 06:30 and 07:30 AM. The observed parameters included Air Quality Index (AQI), temperature, humidity, and air pollutant concentrations. The research stages consisted of sensor data acquisition, data preprocessing using Min-Max normalization, time-series dataset construction using a sliding window approach, and LSTM model training for air quality forecasting. Experimental results showed that the LSTM model was capable of predicting air quality effectively based on temporal sensor data patterns. The evaluation results produced a Mean Absolute Error (MAE) value of 2.046 and a Root Mean Square Error (RMSE) value of 2.076. The findings demonstrate that the integration of IoT and LSTM is effective for real-time air quality monitoring and forecasting. The novelty of this study lies in the use of real-time sensor data collected from a high-traffic road intersection and the integration of monitoring and forecasting systems within a single deep learning-based platform. The proposed system has the potential to support smart environmental monitoring and early warning systems in urban areas.]]>
Copyrights © 2025
