<![CDATA[Accurately predicting the air quality index significantly reduces health risks and supports urban environmental planning. This paper presents LRX, a hybrid predictive model, for Air Quality Index (AQI) prediction. The model employs Long short-term memory to capture temporal dependencies, Random Forest to fine-tune the features, and Extreme Gradient Boosting to enhance the final predictions. The objective of the study is to build a model that can accurately predict air quality index numbers in real time for many cities in India. The proposed model LRX design influences the depth of each algorithm to enhance accuracy and generalization. The experimental results show the model's ability to predict the AQI forecast of various cities in India with a root mean square error of 0.014 and R2 of 0.948, performing better compared to the models individually. To enhance this, a Stream lit-based user interface has been developed to enable real-time AQI predictions and visualization. The interface incorporates tabs for interactive inputs, model selection, graphical representation of predicted trends, ensuring accessibility and usability, and enhancing the practical applicability of the proposed model. This easy-to-navigate tool not only makes the prediction process more accessible but also helps bridge the gap between complex model results and practical environmental decision-making, enhancing the overall impact of the research. This research contributes to air quality prediction by presenting a robust modelling approach that can be applied in the real world.]]>
