previous arrow
next arrow

Languages supported by StarLinker

7 and counting

Adaptive Knowledge and Data Integration

A knowledge and data integration engine for the integration of multi-lingual and multi-domain knowledge and data sources.

Vision and Mission

Due to mobile computing, IoT, social networks, and the overall digitisation of life, humanity is producing ever greater masses of data. Making sense of such a diversity of data is one of the greatest computational challenges today. It is becoming increasingly clear that the bottleneck of automated data processing is not merely volume or velocity but the ability correctly to interpret data within the (often implicit) context of its use and in combination with other data. In computing terms, this transpires as a necessary step of data integration before any insights can be gained through data analysis.

StarLinker is our ongoing effort towards a comprehensive data integration solution that tackles this issue. The ‘magic ingredient’ of StarLinker is adaptivity that we understand as the ability to disambiguate and interpret data with respect to its context: location, time, but also language and application domain (education, healthcare, finance, etc.). The context is a parameter of the system, materialised as pluggable knowledge that includes, among others, language-specific lexicons, domain terminologies, data schemas, or grammars for specialised languages. Adaptation of data integration to a new or changing environment thus becomes as simple as updating the pluggable knowledge in real time.

The quality of data integration ultimately depends on the underlying knowledge being accurate and up-to-date. In an open, diverse, and evolving world, the only way to maintain knowledge at a high level of quality is through the purpose-driven, bottom-up effort of local experts such data scientists, domain experts, or linguists. Through the framework of StarLinker, KiDF provides the opportunity for such experts to collaborate on the development and maintenance of reusable local knowledge, in the objective of solving complex problems of data integration and interoperability.

Download Vision and Mission

StarLinker Statistics

Partner Institutions

Our Project Types

Knowledge and Data Integration

Integration of structured data in multiple languages and domains, as well as the underlying linguistic and ontological knowledge.

View Projects

Multilingual NLP for Data Integration

R&D on language-specific NLP components and pipelines for knowledge extraction from text in structured data.

View Projects

Ontology Matching

Automated matching of ontologies, data schemas, and classifications, expressed in multiple languages and pertaining to multiple domains.

View Projects