Ensuring adequate indoor conditions is essential for healthy and sustainable spaces. However, most studies focus on single environmental factors, overlooking personal and contextual variables. This research conducted a multi-seasonal campaign in an office environment, collecting environmental data (temperature and humidity), personal information, and comfort feedback using commercial and portable sensors. A novel predictive model was developed, integrating environmental conditions with subjective parameters such as emotional state, clothing, and physical activity. By incorporating individual subjective responses regarding environmental perception and personal state, the model improves prediction accuracy. This user-centered, data-driven approach supports energy-efficient building management and enables real-time adaptive HVAC systems that dynamically optimize both performance and occupant comfort.