3. Objectives and issues
TourinFlux project
> Providing the tourism industry with a set of tools allowing them :
– to handle both their internal data, and the information available
on the Web,
– to generate dashboards,
Présentation3
5. Objectives and issues
TourinFlux project
> Providing the tourism industry with a set of tools allowing them :
– to handle both their internal data, and the information available
on the Web,
– to generate dashboards,
– to publish dashboards on the Web,
– to improve the display of information available about their
territory on the Web.
Allow them to improve their decision process and their
effectiveness.
Présentation5
6. Objectives and issues
Présentation6
Problematic
> Tools used currently are inadequate to meet these needs.
> Problem of collecting and exchanging tourism data.
> Problem of diffusing tourism data on the Web.
Locks
> Facilitate collecting data on the Web.
> Facilitate data exchange between different tourism’ actors :
Interoperability.
> The standardization of tourism data modeling :
Homogeneity.
> Facilitate the publication of these data on the Web.
8. Tourism standard : TourInFRANCE
What?
> TourInFrance (TIF): standard established in 1999 by Atout France.
> Used today by more than 3000 tourism’ actors in France.
> Facilitate data exchange between these different actors at a national level.
> In 2004, the TIF Technical Group approved the new version of TIF: TIF V3.
Problematic
> Since 2005, the standard has stopped evolving.
> Tourism’ actors adapted the standard to their own needs.
Présentation8
10. Tourism standard : TourInFRANCE
Problematic
> Tourism’ actors adapted the standard to their own needs :
Tourism Information Systems (TIS) not interoperable among themselves.
The exploitation of tourism information is trapped in its own territory.
Impossibility to aggregate these information.
> Language TIF : XML.
Structured data : lack of semantics.
Lack of expressiveness.
Explicit data : impossibility to infer new knowledges.
Présentation10
11. Tourism standard : TourInFRANCE
Présentation11
Solution
> Evolve the TIF standard.
> Applying the concept of ontology to represent the standard terminology.
TIF TIFSem
> TIFSem, an ontology to globally describe tourism domain :
to annotate information sources on tourism,
to support heterogeneous contents,
to facilitate sharing the knowledge it represents,
to ensure data interoperability.
12. What?
> Initiative launched by Bing, Google, Yahoo! and Yandex in 2011.
> Creating a common model for the content of Web pages.
Why?
> Facilitate searching and integrating data on the Web.
Schema.org
Présentation12
18. Université de La Rochelle
L’Université nouvelle génération
23 avenue Albert Einstein - BP 33060 - 17031 La Rochelle
www.univ-larochelle.fr
fayrouz.soualah-alila@univ-lr.fr
Editor's Notes
Hello, my name is … and I’m a researcher working on the project TourinFlux in the University of La Rochelle in France.
The title of this talk is …
In this talk I will present briefly some preliminary results of TourinFlux project.
Particularly I will present my contribution in the project for developing an ontology based model for tourism domain.
First I will start by presenting the TourinFlux project and exposing the …of our project.
So, this work takes place under the TourinFlux project which aims at …
(1) …, and for this collection of data they have to generate (2) …
I have here putted an example of dashboard that can be generated.
* A dashboard is a document regrouping a collection of information describing the tourism activity of a territory.
In this example of dashboard we have some statistics about opinions given by visitors about hotels in a region of France called Rouen.
After generating these dashboards tourism industry actors need to publish them on Web (3) …
This process should allow tourism’ actors to improve their …
But currently the …, we encounter problems for collecting data from different resources and exchanging them with other tourism’ actors.
We encounter also problems of …
Therefore we are working on (1) …, (2) … (we talk about homogeneity), (3) … (we talk about interoperability), and finally (4) …
To achieve these objectives we focused first on tourism standards to model touristic data.
The standard we are interested on is the … called also TIF.
It has been … and is today … to …
But since …
(mille thousand)
To illustrate this evolution of TIF we show in the following example how the standard was transformed.
The figures 1and 2 show different syntaxes of TIF to describe an address ; data represented in the figure 1 respect the standard TIF V3, while data represented in the figure 2 is structured under a derived version of TIF (different tags syntaxes, new tags added, etc.).
But since …
As a result, tourism professionals have adapted the standard to their own needs and proposed their own evolution in an unorganized way, producing TIS that are not really interoperable among themselves and that cannot directly be shared using international standard.
Accordingly, the exploitation of tourism information is trapped in its own territory, and so it is impossible to aggregate these information.
The second problem that we encounter with TIF is that is presented with XML : Structured data in TIF lack of semantics and expressivity, and we can not infer on them to generate new knowledge.
To remedy these gaps we propose to … by …
We evolved TIF to TIFSem, for semantic TIF.
TIFSem will be used to (1) …, (2) …, (3) …, (4) …, and finally to …
The second standard we are interested on is …
It’s an …
The objective is to … to …
As showmen in this simplifier overview of the model, schema.org regroups a list of terms, relations, attributes and datatypes to describe a content on the Web.
As showmen in this simplifier overview of the model, schema.org regroups a list of terms, relations, attributes and datatypes to describe a content on the Web.
We can use these terms, by integrating them on contents to have a better visibility on the Web.
In our approach we want to spread enriched semantic tourist data that can be easily indexed by search engines.
One solution would be to use directly the Schema.org ontology which provides good indexing by search engines and which is easy to implement.
The disadvantages of this solution is that we lose precision in the tourist data of the TIF model, in addition, using directly Schema.org can cause economic problems because a large number of tourist offices already use TIF or derived versions.
A second solution would be to realize an ontology by matching terms of TIF with terms of Schema.org by using OWL relations, and work with Schema.org community to extend the schema, either formally by adding new terms or informally by defining how Schema.org can be combined with some additional vocabulary terms.
To have an overview of our …
In order to elaborate the TIFSem ontology, we have consulted different kinds of sources to enable the understanding and the collecting of concepts related to the specialized domain of tourism, and the corresponding vocabulary.
Sources coming from Departmental Tourism Committee of the Charente-Maritime (CDT17) and Departmental Tourism Committee of the Aube (CDT10) were consulted.
After collecting the concepts and corresponding lexical items from the sources, we started structuring the first version of the ontology.
The mapping with Schema.org complements the ontology to make it more complete, up to date and coherent.
We have also collected information from the Web and from e-tourism Web sites, annotated them, and extracted information about time and opinion to complete the description of touristic objects.
All these information where integrated in a triple store corresponding to the TIFSem model.
The last step is to interrogate the triple store either to generate dashboards or generate personalized touristic tours.