Year: 2025
Type: article
Abstract: The rapid expansion of digital tourism platforms has increased the demand for intelligent systems capable of generating personalized and structured travel itineraries. Traditional planning requires users to manually compile information from dispersed sources, which often results in inefficiencies, planning inconsistencies, and cognitive overload. To reduce human effort and enhance personalization, this research introduces an AI- powered travel itinerary planner that integrates Large Language Models (LLMs) with a modern full-stack architecture combining Next.js, TypeScript, Convex, and cloud-based authentication. The proposed platform allows users to specify destination preferences, duration of stay, budget constraints, and individual interests. The system then produces a detailed and coherent itinerary that includes day- wise activity distribution, descriptions of attractions, approximate travel time, geolocation coordinates, and recommended routes. Through dynamic server-side rendering, real- time backend synchronization, and interactive map visualization, the platform delivers a seamless and efficient planning experience. Experimental evaluation through user studies revealed significant improvement in planning efficiency and user satisfaction. The findings demonstrate the potential of combining AI-generated content with scalable web technologies to reinvent the future of automated travel planning.
Language: English
Topic: Mobile and Web Applications
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W4416537402