With the growing integration of machine learning techniques into geophysical research, their application to ionospheric modeling specifically in predicting Total Electron Content (TEC) using GNSS data, has gained significant traction. While various studies have explored this approach across different global regions, the Philippine sector remains largely unexamined despite its scientific relevance. Situated in the low-latitude ionospheric region, the Philippines experiences complex phenomena such as the Equatorial Ionization Anomaly, making it an ideal candidate for focused TEC modeling. This study presents a machine learning-based approach to predict regional ionospheric TEC using GNSS data from the PIMO receiver station (14.6°N, 121.1°E), covering the period from 2010 to 2020. Three ML algorithms—Random Forest, Support Vector Machines, and Gradient Boosting—are used to develop predictive models using features such as temporal parameters and space weather indices: average interplanetary magnetic field magnitude, Bz-component, solar wind proton density, plasma speed, flow pressure, Kp-index, Dst-index, F10.7 solar flux, AE-index, and the Lyman-alpha index. Model performance are evaluated and compared across the three algorithms, with further analysis conducted on feature importance and dimensionality reduction using Principal Component Analysis. Preliminary expectations suggest that all models will yield predictions closely aligned with observed TEC values, with the F10.7 solar flux and Lyman-alpha indices emerging as the most influential predictors. This work lays the foundation for more comprehensive TEC modeling in the Philippine region by enabling future expansion to include data from additional local GNSS stations, ultimately enhancing the spatial resolution and robustness of regional ionospheric models.