A New Combinatorial Algorithm for Effective Itinerary Planning

Selecting and creating an efficient and optimal trip plan is the most aggravating job and a knapsack problem. Although existing recommender systems can provide some predefined travel packages, they are not customized for each specific customer according to their POI. Past endeavors address the issue by giving an automatic travel planning service which sorts out the points-of-interests (POIs) into a modified travel package. Using this customized recommender system which is elaborated as a hybrid model, the user can get optimal package based on the points selected by them. To address the above limitations, the system provides a hybrid recommendation system with rank based recommender model with the use of self organizing feature maps. The current proposal provides an automatic itinerary suggestion and generation service for the backpack travelers, which results in the knapsack problem. The proposed service creates a customized and effective multiday itinerary based on the user's POI. This handles the famous NP-complete problem, and optimal service selection problem. To obtain the optimal solution, a two-stage scheme is adopted which happened to be SOM and tabu search. This paper proposed genetic algorithm based recommendation, which helps to select optimal package based on the customized POI and attribute given by the user. This greatly improves the searching and filtering performance by leveraging the genetic algorithm. Genetic algorithms remain in the wider class of evolutionary algorithms (EA), which helps to create solutions to optimization problems using techniques inspired by natural evolution.