RERS: A real‑time emotion‑aware recommender system for sustainable and eco‑friendly driving

Today’s ever-increasing urbanization and mobility demands highlight the importance of sustainable transportation systems. Driving behaviors and habits, which are closely linked to emotional states, also impact road safety, fuel efficiency, and environmental sustainability. In this context, the proposed system aims to contribute to sustainable urban mobility by offering real-time emotion recognition and recommendations for drivers. The proposed system is a pipeline that utilizes physiological sensors, embedded systems, a Web API, a mobile application, a pre-trained machine learning model, and a recommendation agent. The wearable device equipped with physiological sensors sends the acquired data to the Web API to be classified with the machine learning model. The classified emotions and generated recommendations can be monitored live through the developed mobile application, and user-centered emotion tracking is provided by analyzing the retrospective data. To evaluate the system’s accuracy, a series of test sessions were conducted with users, and the overall accuracy of the proposed system was calculated as 75.01%. Designed to be modular, scalable, and adaptable, the system contributes to sustainable urban mobility by encouraging eco-friendly driving practices as well as contributing to driver well-being through real-time emotion recognition and recommendation. It could also serve as a framework for similar systems that can be developed in various contexts.