Neural Collaborative Filtering for Improved Tourism Destination Recommendation

This study evaluates the effectiveness of Neural Collaborative Filtering (NCF) in enhancing tourism destination recommendation. The researchers employed a methodology encompassing a comprehensive literature review, integration of user preference dataset, and the application of the NCF algorithm. The evaluation utilized MAE and RMSE metrics, demonstrating significant improvement compared to the Neural Network method. The results highlight the superior accuracy and reliability of the NCF-based recommendation system in providing tourism destination recommendation. This study emphasizes the practical significance of implementing NCF, offering a more sophisticated and efficient approach to enhancing user experience. The implication of this research contribute to academic needs and are expected to provide practical benefits for recommendation system in the tourism sector. The researchers elucidate critical differences in the NCF algorithm compared to benchmarks, and the introduction explores gaps in NCF applications for tourism, reinforcing the importance of this study. Experimental results show that the NCF model achieved an MAE of 0.306 and an RMSE of 0.371, outperforming the Neural Network's MAE of 0.794 and RMSE of 0.960, confirming superior predictive accuracy.