Machine Learning Approach to Cluster Destination Image on Social Networks

Abstract

 The increasing use of Instagram as a plat form to share travel experiences has led to a vast amount of  visual content related to destination images. This study pro poses a machine learning approach to cluster destination  images on Instagram. The objective is to identify the under lying patterns and themes in travel images shared on Insta gram, which could provide useful insights for the tourism  industry. The study uses a dataset of 10,000 Instagram imag es with destination tags and applies a deep learning approach  to extract visual features from the images. K-means cluster ing is then applied to group images based on visual similari ties. The results show that machine learning techniques can  be used to cluster destination images on Instagram into  meaningful categories such as natural landscapes, cultural  landmarks, food, and cityscapes. These insights can be used  to develop targeted marketing strategies and improve tour ism experiences.