Most travel service providers use XML based schemas to describe their services so that they are amenable to automated discovery by search engines and interpretation by software agents on the Web. For such organisations, the addition of ontological descriptions and semantics is the next logical step that can help to improve the traveller experience and relevance of search results, both within the booking engine and on general purpose search engines like Bing, Google and Yahoo!.
A travel ontology is a conceptualization and formalization of the travel domain using ontology languages such as RDF and OWL. Travel ontologies conceptualise:
Travel processes and the entities that participate in them such as holidays, journeys, people, geographical features, etc.;
Physical and abstract things such as land, countries and regions;
Material and immaterial things such as buildings, vehicles and journey plans;
Attributes of things used for classification purposes such as long journeys, fast vehicles, etc.
Roles / functions of things involved in travel (traveller, booking party, travel agent, etc).
The sample travel ontology as proposed by Chang Choi et al [1], defines an upper ontology consisting of key concepts such as accommodation, activity, food and transportation that can be further specialised. Logical descriptions are further used to define the meaning of the concepts and of their relationships. Currently, only experimental, small scale travel ontologies have been created [2] . They define class hierarchies of concepts such as hotel, restaurant, sports, specialisations such as luxury hotel, bed and breakfast, services and activities such as safari, hiking, spa treatment, sunbathing, sightseeing, accommodation rating (three stars, etc.), campground, surfing, consisting of primitive definable classes, their relations, properties, domains and ranges as well as individual entities.
Travel & Traveller Knowledge Representation - Travel Ontologies
A travel ontology has been proposed by the SWAT Project [3]. The current version (v26) of the ontology consists of 146 classes, 64 object properties, and 8 data properties. GTFS [4] which has been superceeding the Transit vocabulary [5] is a translation of the General Transit Feed Specification [6] as an open standard data format for public transport. TAGA is an agent framework for simulating the global travel market on the web [7]. The framework provides several ontologies: Travel Business; Auction; FIPA OWL content language; TAGA Query language. The travel business ontology contains 34 classes and 16 object properties. Travel Guides [8] utilizes Semantic Web technologies to decrease the maintenance efforts required for existing e-Tourism systems and ease the process of searching for vacation packages. It provides a travel ontology as an extension of the PROTON Ontology, contains 67 classes, 19 object properties and 3 data properties. Knowledge Model for City and Mobility KM4C is a knowledge model to describe a smart city, that interconnect data from infomobility service, Open Data and other source [9]. To summarize the survey on
Travel Ontologies, three important aspects in the travel domain are examined:
1. Capability to describe activity of moving from one location to another.
a. Describe who is travelling.
b. Describe the original location.
c. Describe the final location.
2. Capability to describe means of transportation used when moving from one place to another.
3. Capability to describe activity to stay for a relatively short time in between movements.
An estimation value from 0…1 is assigned to each aspect, where 0 means the aspect does not supported
completely and 1 means highly supported (Table 1).
|
No
|
Ontology Name
|
Aspect #1
|
Aspect #2
|
Aspect #3
|
|
1
|
Travel Ontology of SWAT Project
|
1
|
1
|
0
|
|
2
|
General Transit Feed Specification
|
0
|
0.2
|
0
|
|
3
|
Travel Agent Game in Agent cities
|
0
|
0.2
|
0.2
|
|
4
|
Travel Guides
|
1
|
1
|
1
|
|
5
|
Knowledge Model for City and Mobility
|
0.2
|
1
|
0.4
|
Table 1: Important aspects in travel domain
Schema extension for Travel Domain
Schema.org consists a wide variety of classes and properties [10]. Specifically for travel activity as the movement of people between distant geographical locations [11], we found a highly related class in Schema.org so called “MoveAction”.
“MoveAction” is the act of an agent relocation to a place, where the agent can be “Person” or “Organization”. This action requires two “Place” as original and final location respectively. It consists of three sub-classes:
1. “ArriveAction” as the act of arriving at a place,
2. “DepartAction” as the act of departing from a place,
3. “TravelAction” as the act of traveling from a location to a destination.
“MoveAction” has two specific properties, “fromLocation” depicts the original location and “toLocation” as the final location. “TravelAction” itself has property “distance” depicts the distance travelled. Another important aspect for travel activity is it requires a mean of transportation to be able to move from the original to final location. This mean of transportation might be foot, bicycle, train, airplane, etc.
Semantic enrichment of travel and tourism data and systems - Semantic annotations
Semantic Web technologies are expected to influence the next generation of Destination Management Systems by providing interoperability, reusability, and shareability among modular and service-oriented Destination Management Systems. The Semantic Web is a concept that enables better machine processing of information on the Web, by structuring documents written for the Web in such a way that they become understandable by machines. The Semantic Web allows users and travel agents to make specific queries and infer knowledge quickly and accurately. The major components of the Semantic web framework are the ontologies, the ontology languages, the semantic annotations, the software agents and the applications/services.
The field of using semantics in tourism is not new. The SATINE project [12] by Dogac et al. describes how to deploy semantically enriched travel Web services and how to exploit semantics through Web service registries. Jakkilinki, Georgievski, and Sharda proposed an ontology based e-Tourism Planner AuSTO [13] that enables users to create an itinerary in one single application by this intelligent tool that builds on semantic web technologies.
In the specific field of using semantics to provide better information for travellers there have been relevant and recent efforts in the literature. For example, García Crespo et al. proposed a semantically enriched recommendation platform [14] for tourists on route, later expanded to DMS.
Semantic search for travel portals / travel planners
The introduction of semantic search during the online search phase of a travel booking is becoming a game changer, even though it is still in its infancy regarding the travel industry. Semantic search enables the traveller to focus his search priorities within an open search text box. Now days, a growing number of websites, including travel sites, are moving towards semantic search. The reason is simple: the travel industry does not have much choice. Travel search based on traditional search takes a lot of effort and requires time and patience, and time-consuming solutions are not what internet users are after.
An important issue that is addressed in the context of the semantic search, is the fact that there are a massive number of searches that go on today using online booking engines that are ultimately booked using a call center or reservation service due primarily to the lack of pertinent search data available online. Hotel companies, CRS platforms and OTA’s must now be able to give the travelling public exactly what they want in order to be pertinent. The introduction of semantic search will enable the travel experience to evolve constantly and get continuously better.
Recently, semantic search has become relevant to e-commerce sites, including travel ones. One of the world’s top websites in online travel industry, Expedia.com, is working on the semantic search tools for customers that seek travel products. A beta version of a semantic search service has also been launched on Cheapair.com. Another travel site going semantic is Adioso.com. ZapTravel.com is yet another example of a travel website, which took up the challenge. The creators of the site ensure that their search engine responds to queries in natural language.
[1] ICACT 2006 Conference: Advanced Communication Technology, 2006. ICACT 2006. The 8th International Conference, Volume: 1
[2] View at http://www.owl-ontologies.com/travel.owl#
[3] SWAT Project, View at http://swatproject.org
[4] GTFS Specification, View at http://vocab.gtfs.org/
[5] Transit vocabulary, View at https://github.com/iand/vocab-transit
[6] General Transit Feed Specification, View at https://developers.google.com/transit/gtfs/reference
[7] TAGA Framework, View at http://taga.sourceforge.net/
[8] Travel Guides, View at https://sites.google.com/site/ontotravelguides/Home
[9] KM4C, View at http://lov.okfn.org/dataset/lov/vocabs/km4c
[10] Schema.org, View at http://schema.org
[11] Travel, View at https://en.wikipedia.org/wiki/Travel
[12] Intangible, View at http://schema.org/Intangible
[13] Dogac, A., Kabak, Y., Laleci, G., Sinir, S., Yildiz, A., Kirbas, S., et al. (2004). Semantically enriched Web services for the travel industry. ACM Sigmod Record, 33(3), 21–27.
[14] Jakkilinki, R., Georgievski, M., & Sharda, N. (2007). Connecting destinations with an ontology-based e-tourism planner. In M. Sigala, L. Mich, & J. Murphy (Eds.), Information and communication technologies in tourism (pp. 21–32). Berlin, Germany: Springer Wien.
[15] García-Crespo, A., Chamizo, J., Rivera, I., Mencke, M., Colomo-Palacios, R., & Gómez-Berbís, J. M. (2009). SPETA: Social pervasive e-Tourism advisor. Telematics and Informatics, 26(3), 306–315.