As a highly innovative SME, TILDE participates in national and international research and development projects. In research and development projects co-financed by the European Commission, TILDE has acted as coordinator in five out of the eleven FP5- FP7 and ICT-PSP projects in which it has participated. Tilde was a technology and content provider and research partner in the remaining six projects.
Research projects are related to our main research competences:
Many of the languages not supported by our current technologies show common traits: they are morphologically complex, with free and diverse word order. Often there are not enough training resources and/or processing tools. Together this results in drastic drops in translation quality. The combined challenges of linguistic phenomena and resource scenarios have created a large and under-explored grey area in the language technology map of European languages. Combining support from key stakeholders, QT21 addresses this grey area developing:
- substantially improved statistical and machine-learning based translation models for challenging languages and resource scenarios;
- improved evaluation and continuous learning from mistakes, guided by a systematic analysis of quality barriers, informed by human translators;
- all with a strong focus on scalability, to ensure that learning and decoding with these models is efficient and that reliance on data (annotated or not) is minimised. Project website
FREME addresses the general systemic and technological challenges to validate that multilingual and semantic technologies are ready for their integration in real life business cases in innovative way. These technologies are capable to process (harvest and analyse) content, capture datasets, and add value throughout content and data value chains across sectors, countries, and languages. Project website
The MLi Support Action is working to deliver the strategic vision and operational specifications needed for building a comprehensive European MultiLingual data & services Infrastructure, along with a multiannual plan for its development and deployment, and foster multi-stakeholders alliances ensuring its long term sustainability. Project website
The project results is a web based news service consisting of the real time social sentiment about a set of financial products.
The news are multilingual (Latvian, Swedish, German, Dutch, Polish, French) social media sources (blogs, feeds). Tilde ensured multilingual social media translation for social sentiment analysis by matured and specially adapted for social networks and financial domains SMT (statistical machine translation) systems. The news feed will be available as a free version listing the sentiment only, and a paid subscription based feed offering added services (links to originating news message, personalization and archive functionality). Project description
The TaaS project addressed the need for instant access to the most up-to-date terms, user participation in the acquisition and sharing of multilingual terminological data, and efficient solutions for terminology resources reuse. The developed cloud-based TaaS platform provides the following online core terminology services:
- Automatic extraction of monolingual term candidates from user uploaded documents using the state-of-the-art terminology extraction techniques
- Automatic recognition of translation equivalents for the extracted terms in user-defined target language(s) from different public and industry terminology databases
- Automatic acquisition of translation equivalents for terms not found in term banks from parallel/comparable web data using the state-of-the-art terminology extraction and bilingual terminology alignment methods (MS2: Prototype bilingual term extraction system/M12)
- Facilities for cleaning up (i.e., revising: editing, deleting) of automatically acquired terminology by users
- Facilities for terminology sharing and reusing: APIs and export tools for sharing resulting terminological data with major term banks and reuse in different user applications (MS3: TaaS platform and integrated core services). Project website
To fully exploit the huge potential of existing open SMT technologies the project proposed to build an innovative online collaborative platform for data sharing and MT building. This platform supports upload of public as well as proprietary MT training data and building of multiple MT systems, public or proprietary, by combining and prioritizing this data. Project website
The aim of the ACCURAT project was to research methods and techniques to overcome one of the central problems of machine translation (MT) – the lack of linguistic resources for under-resourced areas of machine translation. The main goal was to find, analyze and evaluate novel methods that exploit comparable corpora on order to compensate for the shortage of linguistic resources, and ultimately to significantly improve MT quality for under-resourced languages and narrow domains. Project website
The META-NORD project aimed to establish an open linguistic infrastructure in the Baltic and Nordic countries to serve the needs of the industry and research communities. The project focused on 8 European languages - Danish, Estonian, Finnish, Icelandic, Latvian, Lithuanian, Norwegian and Swedish - that each have less than 10 million speakers. The project assembled, linked across languages, and made widely available language resources of different types used by different categories of user communities in academia and industry to create products and applications that facilitate linguistic diversity in the EU. Project website
The TTC project aimed at leveraging machine translation tools (MT tools), computer-assisted translation tools (CAT tools) and multilingual content management tools by automatically generating bilingual terminologies from comparable corpora in several European languages (i.e. English, French, German and Latvian) as well as in Chinese and Russian. Terms in different languages are aligned based on the similarity of words next to them in the corpora (immediate vicinity), the approach is known as lexical context analysis. The system generates candidate translations for single- or multi- word terms. The approach relies on the one-to-one relation between terms and concepts.
The project supports the e-inclusion of disabled and elderly people, by providing crosslingual and multimodal support for accessing information bases on assistive tools and technology. Recent efforts have linked national assistive technology information bases into a European portal called EASTIN . The objective of EASTIN-CL was to enhance this portal by creating a front-end to make it more accessible, using language technology:
- Multilingual technology allowing users to search the data in their native language;
- Multimodal technology allowing them to access the portal not just in written but also in spoken communication.
Tripod project aimed to automatically build rich multi-faceted text and semantic descriptions of the landscape and permanent man-made features pictured in a photograph; and to create a more advanced image search engine. Tripod augmented images with spatial data to compute contextual information about the location and features of the actual landscape pictured. Using 3D models, buildings and landscape features contained in the image are identified and located within the picture. Techniques from Web search and text summarisation were applied to automatically create textual descriptions of the photographs, producing a rich readable and multifaceted caption far removed from merely location but encompassing culturally encoded notions such as socially connoted language of place such as suburb, west end, etc.
The project aimed to investigate and develop techniques to analyse the multi-modal behaviour of users within the context of real applications. The multi-modal behaviour takes the form of eye gaze/fixation, eye blink and body move. The techniques was developed and validated within the context of three different application domains: Security, Customized marketing, and Interactive web TV.
The aim of the CLARITY project was to develop cross-lingual information retrieval (CLIR) techniques for English -> Finnish, Swedish, Latvian & Lithuanian i.e low density languages with minimal translation resources and to investigate techniques of document organisation and presentation in concept hierarchies and by document genres and filters. Clarity was a fully-fledged retrieval system that supported the user during the whole process of query formulation, text retrieval and document browsing.
The EuroTermBank project focused on harmonisation and consolidation of terminology work in new EU member states, transferring experience from other European Union terminology networks and accumulating competencies and efforts of the accessed countries. The EuroTermBank project result in a centralized online terminology bank for languages of new EU member countries interlinked to other terminology banks and resources. Project website