Dengue is a severe public health issue in the Philippines with no specific treatment regime, leading to a focus on early detection. Meteorological data, temporal data, and search engine trends have been shown to be accurate predictors of dengue; however, no known system utilizes all of these data points. As such, this study aims to develop a machine learning model for the real-time prediction of dengue in Quezon City, Philippines. A Long-Short Term Memory (LSTM) neural network was trained and evaluated on datasets containing daily rainfall, temperature and humidity sourced from the Philippine Atmospheric, Geophysical and Astronomical Services Administration (PAGASA), dengue case data from the Quezon City Health Department’s Epidemiology and Surveillance Division (QCHD-ESD), and search engine trends from 2010 to 2024. Recent tests showed that the model performed most accurately when trained at 100 epochs and a learning rate of 0.01. This resulted in, at best, a root mean square error (RMSE) of 5.26 when evaluated on internal data and an RMSE of 17.4 when forecasted values were evaluated on external data from January 2025. The model’s forecast accurately showed that an outbreak would occur around 2 weeks in advance, indicating that the model is capable of predicting dengue outbreaks in Quezon City. A prototype website displaying the model’s forecasts was also developed, including functions to retrieve meteorological data from the past 24 hours and provide forecasts for the next 30 days. This study contributes to the currently existing preventative efforts against dengue in the Philippines.