Air pollution, one of the major concerns for the environment and public health concern, requires accurate monitoring and prediction tools. This research presents an innovative air quality index (AQI) prediction model, utilizing a Random Forest algorithm specifying the dynamic and spatial unpredictability regarding air pollutants. Our approach enhances the personalization of predictions by integrating the user's health history and activity data. Using a wide range of datasets from aqi.org, enclosing three months of air quality updates ensures appropriateness and relevance. The model’s explainability is achieved using the SHAP Python library providing explanations through multiple graphical representations. This not only helps in understanding the influence of various factors on AQI but also increases the trust in the model’s predictions. With enhanced precision, our model displays the effectiveness of the results with a high accuracy rate of 90%. Our improvement regarding this topic lies in the integration of user-specific data with advanced machine learning techniques to present personalized air quality monitoring.