Trpkovski, Aleksandar (2018) Tourist Interest Mining From Online Hotel Photos. Research Master thesis, Victoria University.

Abstract

The Internet has been serving as an effective marketing tool in tourism. It provides both the business and the customers with a valuable tool for information sharing, communication, and online purchasing. Photos posted along traveling are fast and up to date, and as they are available everywhere, and they have become the word of mouth of the digital age. Research accomplished throughout the last couple of decades has demonstrated that the photo is an essential concept for understanding the process of selection of a destination by tourists. They are easily recognised and remembered by customer than words. Photographs were also found to attract attention from a customer more than textual content when visiting web sites of online shops and they evoke feeling and generate desire for the associated products or places. However, there have been very limited attempt to assess visual content and visual quality of online photos. That is probably due to the limited background in photography among tourism researchers, and the inefficiency of manual assessment approach for large scale dataset such as the case of online hotel photos shared by travellers. As a result, prior works were unable to provide comprehensive understanding about traveler's perception and interests toward hotels, especially the differences between visual photo content and visual photo quality shared by different groups of travellers online. Aiming to overcome these barriers, the focus of this thesis is to introduce computational approaches to automatic photo content recognition and visual features extraction for automatic photo quality assessment from online hotel photos.

Search Google Scholar

Repository staff login