ExplainableTrip: An XAI-Driven Personalized Travel Recommendation System Using Gradient Boosting and SHAP

Abstract

Tourism is an essential sector of the world's economies, necessitating to have travel planning solutions that are personalised and efficient. This work therefore presents ExplainableTrip, an advanced travel recommendation system utilizing Explainable Artificial Intelligence (XAI) to provide expedient and interpretable trip suggestions. With different user preferences, travel constraints, and comprehensive point-of-interest data, generates highly accurate itineraries, achieving an overall accuracy of 98.49% and an F1-score of 0.97. ExplainableTrip leverages SHAP values (SHapley Additive exPlanations), to explain how factors like travel distance, time constraints, user interests, and Point of Interest (POI) popularity influence final recommendations. This will enable users to understand why a given set of options is being suggested, thus trusting the system. The methodology used in the system is an elaborate study of POI and inputs from the users based on Jaipur, India, alongside advanced algorithms for distance computation, time slots allotment, and route optimization. SHAP analysis forms a core feature, which enables users to receive customized itineraries while understanding the rationale behind them. ExplainableTrip is an enhancement of travel planning with balance between AI-driven efficiency and interpretability, and combines the latest technology with user-centric transparency to meet contemporary travellers' need for personalization and clarity in trip planning.