Using a complex combination of cutting-edge machine learning algorithms and a wide range of meteorological factors, this research offers a unique method for forecasting dengue cases. Our model uses complex techniques including Support Vector Regression (SVR), Gradient Boosting, Conv1D, and the powerful attention model. It leverages a large dataset with temperature, humidity, rainfall, wind speed, surface pressure, and historical dengue cases. The approach improves forecasting accuracy by including spatiotemporal data, which allows for proactive dengue prevention and control via resource allocation and appropriate actions. Thorough assessment confirms the model’s effectiveness and establishes its dependability for public health authorities. Notably, the model serves as a proactive tool for early response measures when the attention model is functioning well, considerably lowering the rates of dengue illness and mortality. Among all the models, the attention-based mechanism model performs well with an RMSE: 0.057. Crucially, this study emphasizes how important it is to use state-of-the-art machine learning techniques to improve predictions in the ever changing field of infectious diseases. It is important to remember that the West Godavari district provided the carefully gathered data for this research, which offers a relevant and specific viewpoint on dengue dynamics in the area.