| Abstract |
Ride-sharing services have become a favorite mode of transportation, permitting multiple passengers to share a ride and cut down the vehicles count running on the roadways. This approach not only decrements transportation costs but also reduces traffic congestion and pollution, making it an excellent use-case of a smart city application. In the context of ride sharing, quality of experience (QoE) is a critical parameter that evaluates users' satisfaction with their ride-sharing experience. To enhance QoE, we propose a recommendation system that integrates user profile details from online social networks (OSN) with customer liking. Our goal is to upgrade users' QoE by creating a ride-sharing profile for each user based on their personality and preferences. To achieve this, we conduct subjective tests to gather users' preferences and evaluate the results using machine learning algorithms to construct user profiles. The experimentation demonstrates that our approach is effective, with a Random Forest classifier, attaining an accuracy of 95%, precision of 91%, and recall of 96%. These results indicate that our system can accurately identify users with similar personalities and preferences, resulting in a more personalized ride-sharing experience. © 2024 IEEE. |
