Roman Egger built up the tourism research department at the Salzburg University of Applied Sciences, of which he was head until 2010. He relinquished the leadership in order to be able to devote more time to his personal research interests. As a tourism researcher Egger is internationally aknowledged, his numerous scientific publications as well as the large number of conference contributions are proof of this. Recently, Egger has also increasingly devoted himself to topics that go beyond the tourism aspect. At present AI, Data Science and methodological issues are at the forefront of his work.
Also see the projects I do with my students
Current research foci are:
- Artificial Intelligence
- Computational Social Sciences in Tourism
- Methods of Tourism Research
- Data Science
- Virtual Reality / Augmented Reality in the tourism and leisure sector
- Gamification and Experience Design
- The use of new technologies in the tourism and leisure sector
- Open innovation, co-creation and crowdsourcing
Current research projects:
The impact of Large Language Models such as GPT for the tourism and hospitality industry. I am doing mainly consulting and development projects (e.g. GPT-based Chatbots that allow to Q&A about your own external data)
Travel search behaviour of Austria’s main tourism markets
In the last few months, dashboards were developed for the Austrian national tourism office, Vienna Tourism and Tyrol to visualise the development of Google search queries in the context of travel. For this purpose, a complex procedure was developed (you can view the project report here) to enable valid comparison of search results. The live dashboard of the Austrian National Tourist Office can be viewed at: https://www.austriatourism.com/oew-global/oew-global-dashboard/ (link Google search results). If your destination is also interested in such a dashboard, please contact me.
Data from guests‘ mobile devices is considered promising for modeling tourism behavior and visualizing visitor flows. Especially in times of overtourism, the analysis of movement data is of great interest and visitor guidance is considered the buzzword of the day. I am currently analyzing and visualizing such data in different projects. The questions are: where do the tourists come from, which places, sights and attractions do they visit and are there differences in nationalities. How long do tourists stay at these places and are there typical movement patterns in the destination?
The image below shows moving patterns from German Tourists during summer of 2020 in Salzburg. Database is about 1 Mio Datapoints.
Creating a tourist typology based on images
Tourists are multi-optional. One time they go to Croatia by camping bus, the other time to a 4* hotel in Tyrol for skiing. Once a shopping trip to London, then pilgrimage on the Way of St. James. How can such hybrid profiles be typologised and characterised? To solve this problem, a method was developed in which people select pictures (behind which are annotations that have been converted into a 100-dimensional vector, clustered with Louvain’s algorithm. This is followed by a dimensional reduction t-SNE or UMAP and a visualisation).
Um sich selbst zu „verorten“ können Sie das Tool hier ausprobieren.
Instagram – Cluster Map
Around 100,000 Instagram posts about Austria were analysed for the Austrian National Tourism Office. For this purpose, the images were crawled and analysed for hashtags specific to Austria. Clusters were created (multiple layers of k-Means) and visualised on a map in the dashboard. This allows the perceived customer perspective to be displayed geographically, which enables the destination to discover „blind spots“ in destination marketing. The paper „A Machine Learning Approach to Cluster Destination Image on Instagram“, will soon be published in the journal „Tourism Management“ (open access).