DETECTING NOVELTY SEEKING FROM ONLINE TRAVEL REVIEWS A DEEP LEARNING APPROACH

Abstract

In recent years, online travel reviews have become a valuable source of information for understanding tourist behavior and preferences. This study focuses on detecting novelty-seeking tendencies in travelers by analyzing online travel reviews using a deep learning approach. Noveltyseeking, characterized by a desire for unique, new, or unconventional experiences, is an important factor in tourism that influences destination choice, activities, and overall travel satisfaction. Traditional methods of understanding this behavior have relied on surveys and qualitative analysis, which can be time-consuming and limited in scope. To address this, we propose a deep learning-based model that automatically identifies novelty-seeking behaviors from unstructured online travel reviews. The model is designed to capture key linguistic and semantic patterns in reviews that suggest a traveler’s inclination towards exploring new and unconventional experiences. By leveraging advanced natural language processing (NLP) techniques and deep neural networks, the model processes vast amounts of review data, learning to detect novelty-seeking traits based on travelers' descriptions of their experiences. Our approach demonstrates high accuracy in predicting novelty-seeking tendencies, outperforming traditional machine learning methods. The findings reveal important insights into how novelty-seeking influences travel decisions and how destinations can tailor their offerings to attract these travelers. Furthermore, this automated system can provide tourism businesses with a scalable tool to analyze customer preferences, helping them improve marketing strategies and enhance the travel experience for novelty-seeking individuals.