Promotion of tourist attractions and recommendations to visitors with open data

Share

Fecha de la noticia: 15-02-2022

As our lives become more and more digitized, activities as face-to-face as “tourism” are also being pushed towards a transformation as profound as that of other sectors and activities. In this digitization process, both the data and the technologies associated with artificial intelligence are essential and this was highlighted, for example, by the 2020 European Tourism Convention.

The importance of tourism for the Spanish economy is enormous (12.4% of GDP and 12.7% of employment in 2019). The number of visitors we receive in our country, even despite the break caused by the pandemic, is still among the highest in the world. Therefore, in all the strategies and plans that are developed in our country, the weight of the tourism industry is reflected. For example, with the deployment of the Spanish Hub of Gaia-X, whose objective is to make it easier for sectorial industries of all sizes to create community and promote innovation based on data and Artificial Intelligence, Spain aspires to become the first country to lead a data space in the tourism industry. Bearing in mind that the hubs must facilitate and support the creation of European data spaces and are representatives of the regional economy, significant efforts will be made in Spain to develop data spaces that are relevant or specific to the Spanish economy, as is the case of the tourism industry.

As visitors or tourists we all have the expectation of enjoying an increasingly personalized experience, not only during the pre-planning process, but also during the time we are traveling or carrying out the activity. In this sense, open data plays a central role, both to help us select activities or attractions and to obtain relevant recommendations in real time that help us enjoy the experience more. For this reason, a multitude of cities and regions have been betting on publishing data sets and, in some cases, applications specifically aimed at visitors. Serve as an example the almost 3,000 data sets that can be found in the tourism category of the datos.gob.es portal, and which come both from the state administration and from different local and regional administrations.

Construction of aggregate datasets

However, as in other sectors of activity, to develop more sophisticated use cases related to tourism, taking advantage of the advanced use of data and artificial intelligence, data sets that transcend local spheres are necessary. Open data on tourism, as it happens in other domains, is distributed through different websites and in different formats or data structures. In this sense, there are some examples of transnational projects such as Tourpedia what prentend to build and maintain open data sets of high potential value for the development of new use cases and that are now only within the reach of large tour operators.

Tourpedia, which has been developed within the framework of a European project in which Spain, France and Italy participate, lays the foundations for building a single access point for all Italian, French and Spanish open data on tourist accommodation, points of interest and attractions. It also offers a simple mechanism for the integration of new open data sources, with the aim of aggregating open data provided by the public sector from anywhere in the world.

The case of Tourpedia, although it has not reached its full potential, is frequently cited and has been the subject of some scientific articles such as “Open data for tourism: the case of Tourpedia”, where the high impact of creating useful aggregate data sets for tourism is highlighted.

In this sense, in France we find Data Tourism, a less ambitious initiative in terms of its geographical scope, but perhaps more consistent in terms of its implementation. The central idea of DataTourisme, the open data portal for tourism of the French government, is equivalent: to centralize the data collected by local and regional tourism authorities, standardize the formats and make the information freely available for reuse. Data producers gain visibility, while companies and public authorities can integrate data of different types and from different sources into applications and algorithms.

DataTourisme currently covers 96 French departments in 14 regions different that have published as open data more than 385,000 points of interest and events. For the aggregation and publication of data, the portal is committed to using linked data and proposes the DataTourism ontology.

In Spain we have Dataestur, a Segittur initiative that contains a selection of the most relevant data on tourism in Spain. The data added in Dataestur come from sources as diverse as INE, Renfe or Segittur itself and are grouped into five categories for download, consultation and study. It also highlights the fact of publishing an API for developers and reusers with a set of methods that allow automating the download of a large part of the portal's data.

Recommender systems

At present, there are not too many aggregate data sets on tourism and much less those published as linked open data, but there is abundant scientific literature that supports this line of action. In this sense, a meta-analysis of 126 scientific articles, selected for their impact, concludes that the use of linked open data to address location-based recommendation and react in real time to the needs of tourists is widespread in the field of tourism.

One of these scientific articles, a 2020 investigation demonstrates that combining a user's location with open data on TripAdvisor ratings, destination closing time, or traffic can greatly increase the quality and accuracy of recommendations. In total, we find six types of jobs, classified according to the use case they focus on:

Independent Point Location Recommendations. Recommend to the user a point of interest around the current location based on their own preferences.
Travel route recommendations. Provide the user with the recommended route and travel itinerary.
Recommendations based on GPS track. They provide recommendations based on past behavior and travel patterns recorded in the form of GPS tracks.
Recommendations based on geotagged media data. They generate recommendations based on the extraction of multimedia data from texts or photos to discover places, context information and user profiles.
Recommendations based on ontologies. They collect datasets and create tourism ontologies for the different recommended purposes, such as a list of points of interest, popularity of locations, travel itinerary and route planning.
Location-based friend recommendations. They aim to use the user's social connections to recommend places based on the preferences of friends.

It seems that we are beginning to see what could be a second great wave of innovation when it comes to the use of data in the tourism industry. Thanks to the generation of aggregate data sets, the use of linked data and the application of artificial intelligence and machine learning techniques, increasingly sophisticated use cases are being generated for the benefit of the tourist experience and the promotion and discovery of destinations Undoubtedly, the new initiatives that have arisen through the national digitization plans and the support of European funds will accelerate the adoption of many innovations that at the moment we see only in the scientific literature.

Content prepared by Jose Luis Marín, Senior Consultant in Data, Strategy, Innovation & Digitalization.

The contents and points of view reflected in this publication are the sole responsibility of its author.