A Novel Way to Browse: Image Travel Wizard

This paper aims at developing an image travel wizard, a hierarchical image guide that can assist people seeking information for travelling. This guide system intends to overcome many of the limitations of the currently available travel information and navigation websites. These traditional websites usually provide some text links with the name of cities and famous resorts for customer search or let users search their interests by inputting a destination name. However these methods require users to have both geographic knowledge and understanding of a specific place already. People sometimes do not have a specific goal about where to go in the mood of a relaxation tour, not to mention to have a name in mind. This image travel wizard provides cognitive images of different categories that are classified accurately into different travel categories in advance. It can lead the users to an ideal place of interest through four iterations. In each iteration, users only need to click one of the images among three to six choices which are represented on the same screen. Once the users choose one image for a specified category, all the images shown in the next iteration will be selected from this category. There is no need for users to know a place name in advance. These iterations can reduce the category scope gradually until the wizard provides users the final recommendation that precisely meets the users' needs.