Location-based Tourism Recommender System using Swarm Intelligence

Abstract

The developments in social networks and internet technologies have drawn researchers from around the world to investigate recommender systems for developing customized location-based services. The overload of information is challenging for people to make decisions and literature opined that little work has been done using swarm intelligence for tourism in Nigeria. The aim is to develop a location-based tourism recommender system using swarm intelligence. The system development approach adopted in this work is the spiral model. The bat, wolf, firefly, and grey algorithms were considered in the development of the recommender model. The data used for analysis was sourced from the TripAdvisor dataset consisting of attraction places, restaurants, and hotels in some southwest states in Nigeria as well as Lagos and Abuja, and visualized using various Python libraries. The results were experimentally validated on the TripAdvisor dataset to show the performance and efficiency of the system. In the comparison of the various algorithms, the recommender model which used the firefly algorithm had mean square error, root mean square error, and mean absolute error that were lower at 0.81262, 0.90145, and 0.66184 respectively. The firefly algorithm was further developed using a collaborative technique and then integrated into a web application that makes recommendations for a tourist. The latter improved the performance of the system when applied to the TripAdvisor dataset. This work will assist tourists in recommending comparable destinations that are most relevant to them.