Skip to main content
Weiter zur Homepage der Europäischen Kommission (öffnet in neuem Fenster)
Deutsch de
CORDIS - Forschungsergebnisse der EU
CORDIS
Inhalt archiviert am 2024-05-27

MAchine Translation systems for the use of Hungarian and Polish Administrations

Ziel

Original research objectives

The MATCHPAD project aims at developing quality machine translation systems between Hungarian and Polish languages on the one hand, from English and to French on the other, which can be used effectively. In particular, the aim is to facilitate the communication between official bodies of the EU Member States with the administrations of these two countries, and to prepare their integration. An additional aim is to stimulate the development of commercial products that may help the industry when addressing foreign markets. MATCHPAD project intends to demonstrate and put on the market 2 new MT systems with Hungarian and Polish languages from English and to French, and so provide access for the administrations, other public bodies and private companies of Europe to information and facilitate contacts between administration, citizen and third parties. The research aspect is particularly important in this project because, if Polish language is somewhat similar to the main EU languages, Hungarian is a very different kind of language. Analysis and Synthesis of the Hungarian are quite different, due to the agglutinative structure of this language far more complicated than the languages developed in the existing MT systems. The Matchpad project is indented to develop a new family of analysis tools for the East European languages.

The main tasks to be completed in cooperation between the software company and the Universities specialised in terminology research, each one in its own language, are:
- Hungarian and Polish Analysis
- Hungarian and Polish terminology
- Hungarian and Polish Syntheses
- Machine Translation Systems assembling and debugging
- Software Localisation Participation of administration institutions in each country assures the high quality of the final product and mass dissemination of the results.

More specifically, the main objectives of the project are:
- To create four operational language pairs prototypes for administration texts with dictionaries of about 25 000 to 30 000 terms
- Hungarian to French
- Polish to French
- English to Hungarian
- English to Polish

Expected deliverables

The objectives of the Matchpad project concerns Multilinguality in digital content and services, as we create Multilanguage tools for Machine translation (and indexation) based on powerful language analysis, transfer and generation (synthesis) modules, actually used by large administrations for European languages and their extension to agglutinative languages will be a very important aspect of our research:
- Localization of Systran MT products in Hungarian and Polish;
- two dictionaries for administration texts Hungarian and Polish terminology (about 50 000 terms) with translations in French and English;
- English to Hungarian and Polish to French Machine Translation prototypes;
- Hungarian and Polish to French Machine translation prototypes;
- Four Machine Translation systems (Hungarian and Polish from English and to French) for administration texts.
The main result of this project is first 4 translation engines belonging to 2 mains categories corresponding to 2 different technical implementation (A and B). It is also as side products, morphological analysers for Hungarian and Polish, and derived technologies such as spell-checking. A.EN based Engines: those engines are based on fully mature EN generic analysis module, relying on a 250 000 source form dictionary. The 2 translation systems obtained (ENPL, and ENHU) inherit the very high quality of the English analysis (developed and shared among all language is quite limited (about 30 000 entries). Both engines reach demonstration level quality. B. PL&HU->FR engines: they are based on a new generation of tranlator, whom development was initiated by this project. They are presumably far more modular and evolutive: however, since they do not any previously developed linguistic component; the obtained quality is still low, besides, the associated dictionaries are also small.

The principal effort has been made on the development of the analysis modules which is the basis of the translation engine and less effort on the synthesis. These engines are still prototypic. C. PL&HU->EN: those engines do not exist, but since most of development for B. was focussed on PL/HU analysis, the costs for producing such a system is not very important (see linguistique development) D. Morphological analysers and spell-checking application.

Wissenschaftliches Gebiet (EuroSciVoc)

CORDIS klassifiziert Projekte mit EuroSciVoc, einer mehrsprachigen Taxonomie der Wissenschaftsbereiche, durch einen halbautomatischen Prozess, der auf Verfahren der Verarbeitung natürlicher Sprache beruht. Siehe: Das European Science Vocabulary.

Sie müssen sich anmelden oder registrieren, um diese Funktion zu nutzen

Programm/Programme

Mehrjährige Finanzierungsprogramme, in denen die Prioritäten der EU für Forschung und Innovation festgelegt sind.

Thema/Themen

Aufforderungen zur Einreichung von Vorschlägen sind nach Themen gegliedert. Ein Thema definiert einen bestimmten Bereich oder ein Gebiet, zu dem Vorschläge eingereicht werden können. Die Beschreibung eines Themas umfasst seinen spezifischen Umfang und die erwarteten Auswirkungen des finanzierten Projekts.

Aufforderung zur Vorschlagseinreichung

Verfahren zur Aufforderung zur Einreichung von Projektvorschlägen mit dem Ziel, eine EU-Finanzierung zu erhalten.

Daten nicht verfügbar

Finanzierungsplan

Finanzierungsregelung (oder „Art der Maßnahme“) innerhalb eines Programms mit gemeinsamen Merkmalen. Sieht folgendes vor: den Umfang der finanzierten Maßnahmen, den Erstattungssatz, spezifische Bewertungskriterien für die Finanzierung und die Verwendung vereinfachter Kostenformen wie Pauschalbeträge.

CSC - Cost-sharing contracts

Koordinator

SYSTRAN SA
EU-Beitrag
Keine Daten
Adresse
1 PARVIS DE LA DEFENSE LA GRANDE ARCHE - PAROI NORD
92044 PARIS LA DEFENSE
Frankreich

Auf der Karte ansehen

Gesamtkosten

Die Gesamtkosten, die dieser Organisation durch die Beteiligung am Projekt entstanden sind, einschließlich der direkten und indirekten Kosten. Dieser Betrag ist Teil des Gesamtbudgets des Projekts.

Keine Daten

Beteiligte (6)

Mein Booklet 0 0