Tourism Recommendation System Using Indistinct Rule-Based Feature Selection and Classification for Tourists

Abstract

The purpose of recommendation systems in the tourism industry is to satisfy visitors by helping individuals make better informed decisions and selecting optimal tours based on daily varying parameters. In the modern era numerous techniques are being used in recommendation systems as a result of an increasing volume of data studies. In this research, real-time unstructured data are gathered from several kinds of travel-related websites, online data providers, and manual data sources. The recommendation system is built around the feature selection method that is described here, and in this work, semantic rules are constructed for unstructured information on the basis of daily-increasing user preferences. The recommendation system sensitive to strong structural assumptions includes proposed embedded feature selection recommendation models. Based on user preferences that continue to evolve regularly, semantic rules are created for the chaotic data sets that are now accessible. In order to increase the accuracy of recommender systems, a fuzzy semantic classification method that combines the fuzzy data classifier approach with semantic analysis is described in this research.