Publication

Abstract : This project aims to develop an advanced hybrid travel recommendation system that leverages emotion extraction from user comments and visual recognition from images to provide personalized travel suggestions. The system integrates three distinct datasets an image dataset, a custom dataset containing location details, sublocation information, user comments, and hashtags, and a user interaction dataset comprising user IDs, locations, and ratings. Emotion extraction is performed exclusively on user comments to capture sentiments, while visual recognition is applied to images to identify and classify visual content.The recommendation system employs a hybrid approach, combining collaborative filtering and content-based filtering techniques. Collaborative filtering utilizes user interaction data to identify similarities between users and recommend locations based on their ratings. Content-based filtering, on the other hand, analyzes user posts, comments, and hashtags to recommend locations and sublocations that match the user’s preferences and interests.To enhance user engagement and accessibility, a user interface is developed, allowing users to input their posts (images and captions) and receive personalized travel recommendations. This interface not only makes the system user-friendly but also facilitates seamless interaction, providing users with insightful travel suggestions based on their unique inputs.Overall, this project demonstrates the effectiveness of integrating emotion extraction and visual recognition into a hybrid recommendation system, offering a sophisticated solution for personalized travel planning and exploration.