Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{196071,
        author = {P.Bhanu Prakash and V. Abhiram and S. Harsha Vardhan and P. Rukmininadh},
        title = {Travel Destination Recommendation Engine Using Machine Learning And Hybrid Recommender Systems},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {1337-1341},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=196071},
        abstract = {This paper presents a novel Resilient Hybrid Multi-Stage Travel Recommendation System designed to address the challenges of information overload, data volatility, and personalization limitations in modern digital tourism. Traditional recommendation systems often rely on static datasets and struggle to adapt to real-time changes in user preferences and travel costs. To overcome these limitations, the proposed system integrates a dual-stage inference pipeline combining K-Nearest Neighbours (KNN) for rapid candidate filtering and a Large Language Model (LLM) for deep contextual reasoning and itinerary generation.
A key contribution of this work is the introduction of a Scraper–LLM hybrid framework, which incorporates real-time cost-of-living data to ensure accurate and dynamic travel planning, achieving a cost prediction error margin below 6%. Additionally, a dynamic profile evolution mechanism blends historical user behaviour with real-time intent, significantly improving recommendation relevance and achieving a 91% match accuracy.
The system demonstrates high efficiency with low latency and minimal resource consumption (~120MB RAM), making it a scalable and cost-effective alternative to traditional API-dependent solutions. Experimental results validate that the proposed hybrid architecture effectively addresses the cold-start problem, reduces data decay, and enhances personalization, providing a robust framework for real-time, budget-aware travel recommendation systems.},
        keywords = {},
        month = {April},
        }