Air quality is a crucial aspect that affects public health and the environment. As public awareness of the importance of air quality increases, fast and accurate information about air conditions becomes essential. This research developed a Telegram bot-based system that not only provides current air quality information but also predicts air quality for the next five days. The system uses real-time data from the OpenWeatherMap API and employs a regression-based prediction model to provide more accurate air quality projections. This bot is designed to provide easy access to information for people, especially in Indonesia, regarding air quality in various cities. The results show that the system has a high reliability level with a 98.5% success rate and 99.9% uptime. The prediction model using Linear Regression shows good performance with an R-squared (R²) value of 0.86, Mean Absolute Error (MAE) of 0.24, and Root Mean Square Error (RMSE) of 0.31. The system also demonstrates optimal response time with an average of 0.83 seconds per request. User evaluation shows a satisfaction level of 4.2/5, ease of use of 4.5/5, and feature completeness of 4.0/5.
Copyrights © 2023