Intelligent tourist attractions recommender system with hybrid AI

The annual rise in tourism increases demand for a robust tourism recommendation system. Big data processing has been adopted as a key component of adaptive recommendation systems, which provide services designed to suit individual users' needs. Big data processing can be applied to tourism recommendations to offer better options for tourists and reduce the distance traveled by users to visit a location of their preference. In this system, Hybrid AI, Llama, and Logistic Regression are used to categorize big data and recommend tourism types that match the user's personality, based on sentiment analysis, distance, and cost of visit. Hybrid AI combines machine learning, which uses statistical models to analyze data, with Llama AI to provide insights and meanings related to locations presented to the user. The system is scalable to handle the complexity of the data collected. It informs users of locations with the lowest Mean Squared Error through cumulative values from different tourism criteria. The current best cumulative RMSE, MAE, and MSE achieved by the system are 0.909, 0.732, and 2.773, respectively, for a training dataset of 438 entries. The MSE is expected to improve with larger datasets for training.