The European Monnet -Multilingual Ontologies for Networked Knowledge- Project (FP7-248458) kicked off on 1 March 2010. Along with other European organizations and higher education institutions, this project is partnered by the Ontology Engineering Group at the Universidad Politécnica de Madrid's Facultad de Informática, led by Asunción Gómez-Pérez.

The goal of the Monnet project is to build multilingual ontologies to retrieve and present information across languages. The Monnet project aims to meet new industry and government needs to better exploit the information available on the Internet.

The integration, aggregation, querying and presentation of multilingual information available on the web are some of the key challenges set by a globalization world, where everyone should be guaranteed information access across language barriers.

The Monnet project will focus mainly on two use cases business intelligence and public sector applications. The proponents of the first Monnet use case are two companies: SAP, an international business software vendor, and XBRL (eXtensible Business Reporting Language), an association promoting the creation and exchange of financial and business information. The key aim of these two Monnet partners is to provide access to business solutions and financial information about businesses across languages.

The proponent of the public sector-related use case is the Dutch company Be Informed, specializing in the application of semantic technologies to information systems development. In Monnet, Be Informed will apply the technologies and methodologies developed as part of the project to applications providing multilingual access to Dutch government information and public services.

Multilingual information

On top of this, international businesses and organizations have for some time been demanding multilingual information management and access solutions. Prominent examples are the United Nations Food and Agriculture Organization (FAO) or the World Health Organization (WHO). Although they specialize in different fields, these organizations both have to offer the information that they generate in the end users' language, as well as manage information submitted in different languages.

The semantic web is, in this respect, an ideal platform for meeting the demands of multilingualism facing these organizations, because the knowledge organization underlying the semantic web can be considered to be language independent. Access from any language could be guaranteed because the knowledge, data, is stored in a language-independent system.

This development would significantly boost participation by developing countries, which have so far actively contributed very little to web construction, because they would be able to access available data in their own language.

Challenges

The major challenges along the road towards this goal revolve around setting up the infrastructures and formalisms for representing and accessing knowledge across languages. The next step would be to get users to publish and map data to their own language. This way, a truly multilingual semantic web will gradually take shape.

The Monnet project sets out to tackle some of these challenges by providing a semantics-based solution for integrated information management and access across language barriers.

In earlier European projects, such as the recent NeOn Project (FP6-027595), the Ontology Engineering Group at the UPM's Facultad de Informática has already helped to develop multilingual ontologies as a solution for managing and accessing the information of multilingual organizations. One of the key contributions of the Ontology Engineering Group to the NeOn project was to develop a linguistic model for associating multilingual linguistic information to domain ontologies (LIR- Linguistic Information Repository), as well as a semi-automatic ontology translation system, LabelTranslator, available as a plug-in to the NeOn Toolkit ontology editor.

Upcoming workshop

One of the first Monnet project dissemination activities was to call the First Workshop on the Multilingual Semantic Web. This workshop, co-located with the International orld Wide Web Conference (WWW2010) to be held in Raleigh, North Carolina, is to take place on 27 April 2010.

The goal of this workshop is to publicize the work of the international community on issues related to new infrastructures, architectures, algorithms, etc., for accessing, retrieving, publishing and visualizing multilingual information on the web. The Ontology Engineering Group at the Universidad Politécnica de Madrid's Facultad de Informática participate in this workshop with the paper “Rivière or Fleuve? Modelling Multilinguality in the Hydrographical Domain”, by researchers Guadalupe Aguado-de-Cea, Asunción Gómez-Pérez, Elena Montiel-Ponsoda and Luis M. Vilches-Blázquez, as stated in the program of the Workshop.

Other Monnet project partners working alongside the School of Computing's Ontology Engineering Group are the University of Bielefeld (Germany), the National University of Ireland and Digital Enterprise Research Institute (DERI-Galway), the German Research Center for Artificial Intelligence (DFKI), the Dutch company Be Informed, the multinational SAP (Germany), and XBRL Europe (Brussels).