An Intelligent Recommendation Method for Tourist Attractions Based on Deep Learning

Area of Science:

• Computer Science
• Artificial Intelligence
• Tourism Technology

Background:

• Increasing global tourism necessitates efficient information access for travelers.
• Traditional methods struggle to meet the dynamic information needs of tourists regarding destinations, accommodations, and attractions.
• Mobile technology offers a convenient platform for accessing travel-related data.

Purpose of the Study:

• To develop an accurate and efficient recommendation system for tourist attractions using deep learning.
• To leverage lightweight deep learning models for mobile accessibility in tourism.
• To improve the travel planning experience by providing personalized destination recommendations.

Main Methods:

• Implementation of the Visual Bayesian Personalized Ranking (VBPR) algorithm, a deep learning approach.
• Development of a lightweight deep learning model optimized for mobile devices.
• Utilizing intelligent systems for data collection and feature extraction relevant to tourist needs.

Main Results:

• The proposed VBPR algorithm demonstrated a high recommendation accuracy of 98.56%.
• The lightweight deep learning model facilitates enhanced access to travel resources via mobile services.
• The system effectively provides personalized recommendations for tourist attractions.

Conclusions:

• Lightweight deep learning models, like the proposed VBPR, are highly effective for tourist recommendation systems.
• Mobile-accessible AI significantly improves the ability of tourists to gather essential travel information.
• Accurate and personalized recommendations enhance the overall tourist experience and address supply-demand challenges.