Tourism Knowledge Discovery in Social Multimedia

Introduction

Tourism has been one of the first domains of human activities, where the role of multimedia has been crucial as a means to capture and archive personal experiences. Recently, the wide adoption of digital cameras and the popularity of social sharing platforms led to the proliferation of travel-centric user-generated content (UGC) on the Web. Currently, there is a wealth of travel-centric user-contributed information and content on the Web that adds significant value to “traditional” sources of tourism knowledge and content (e.g. official travel guides):

• •

  Images and videos captured by casual tourists during their trips and shared through social networks (Flickr, Panoramio, YouTube).

• •

  Blog articles describing the experiences of travelers.

• •

  Structured tourism information, e.g. Wikipedia infoboxes on popular landmarks or areas.

• •

  Tourism behavioral data, e.g. Foursquare check-ins.

• •

  Explicit feedback and travelers’ impressions, e.g. comments in popular travel sites such as TripAdvisor and booking.com.

As a result of the wealth and diversity of online tourism-related information and content, it has been recently possible to employ data mining, multimedia analysis and knowledge discovery approaches on the tourism domain for refining the information about travel destinations, e.g. by enriching the representation of existing Points of Interest (PoIs) and discovering new ones, and for gaining new insights into the visiting patterns and traits of travelers. Such insights have also made possible the development of innovative services that could greatly enhance the travel experience, for instance, PoI recommendation (Popescu & Grefenstette, 2011) or itinerary creation (Tai et al., 2008). Furthermore, tourism knowledge discovery has proved to be invaluable for tourism professionals by providing insights regarding the visiting patterns and preferences of visitors without the need to resort to costly questionnaire-based surveys. This article summarizes several important research problems and novel approaches in the emerging domain of tourism knowledge discovery from social media content and online user activities.

Background

Here, we first motivate the presented work by highlighting the importance of related technologies for the tourism industry as seen in recent research. Then, we describe the main information and content sources involved in the process of multimedia knowledge discovery. Finally, we provide definitions for each of the research problems that are further discussed in the article.