A Hybrid Recommendation Model for Tourist Using Evolutionary Algorithm Combined with Local Search Algorithm for Trip Planning

Recommender systems play a crucial role in assisting tourists with travel recommendations considering the customized demands is a necessary practice for improving the tourism, business, and aids significantly in decision-making process. The study proposes using a evolutionary genetic algorithm (GA) to efficiently determine the shortest tourist route within a constrained area, allowing for quick visits to multiple destinations. The goal is to find the most efficient route with minimal physical exertion. By exploring all possible routes, the GA can identify the quickest path from the starting point to the destination. The study first provides a brief overview of the enhanced GA, followed by a detailed analysis of its construction and solution. The paper introduces a novel approach to recommendation systems in the tourism industry, combining the Genetic algorithm with 2-Opt algorithm for local search adaptation. The hybrid model is used to optimize the system by considering multiple criteria. Data for this study were collected through online google map resources defining a positive ideal distance matrix for 11 locations with 121 edges. Finally, the GA is further optimized by pipeline the 2-opt algorithm, and a simulation is conducted to examine the average path convergence for each individual route by customizing based on the preferences. Finally proposed model is compared against the traditional algorithms such as Heuristic and Graph network algorithms. Subsequently, the hybrid algorithm searches among destinations to recommend the best tourist trip plan to users. Selecting the shortest trip route plan will reduce the travel cost and time. This model enhances the practical value for researchers and for practical application.