Knowledge Integration and Linked Data Technologies
What kind of smart applications can we build, if we were able to integrate all available knowledge, data and language resources in a meaningful way? While Turing's imitation game is exciting, we are focussing on the actual knowledge engineering to build information machines that enable humans to perform more efficiently in their tasks. To achieve this goal, we believe the following two prerequisites must be met:
- Knowledge and Data must be rendered discoverable and then transformed, linked, enriched and integrated homogeneously in a huge semantic knowledge graph (DBpedia)
- Language Technologies must be on the one hand leveraged to understand, categorize and structure available textual content in all its forms. On the other hand, language technology must assist in building adequate interfaces that allow humans to interact effectively with data and information via discovery, querying and reorganization.
Research in this group focusses on contextualising data and ontologies as well as capturing deep linguistic knowledge to improve machine understanding. Besides research, we are taking the extra effort to undertake transfer and innovation projects and support industrial applications such as in the JURION use case.
Main contact: Dr.-Ing. Sebastian Hellmann
Research Areas
- Data Engineering
- Data Integration
- Data-driven Artificial Intelligence
- DBpedia
- Knowledge Engineering
- Language Technology
Projects
- 5. Leipziger Semantic Web Tag (LSWT2013) – Von Big Data zu Smart Data
- ALIGNED – Aligned, Quality-centric Software and Data Engineering
- DBpedia – Querying Wikipedia like a Semantic Database
- DBpedia NIF Dataset – Open, Large-Scale and Multilingual Knowledge Extraction Corpus
- DBpediaDQ – User-driven quality evaluation of DBpedia
- DBpediaDQCrowd – Crowdsourcing DBpedia Quality Assessment
- DL-Learner – a tool for supervised Machine Learning in OWL and Description Logics
- FREME – Open Framework of E-services for Multilingual and Semantic Enrichment of Digital Content
- LIDER – Linked Data as an enabler of cross-media and multilingual content analytics for enterprises across Europe
- LOD2 – Creating Knowledge out of Interlinked Data
- MEX Vocabulary – A Light-Weight Interchange Format for Machine Learning Experiments
- MMoOn – The Multilingual Morpheme Ontology and Language Inventories
- N3 - Collection – N3 - A Collection of Datasets for Named Entity Recognition and Disambiguation in the NLP Interchange Format
- Navigation-induced Knowledge Engineering by Example – a light-weight methodology for low-cost knowledge engineering by a massive user base
- Neural SPARQL Machines – Translating natural language into machine language for data access.
- NIF4OGGD – Natural Language Interchange Format for Open German Governmental Data
- NLP Interchange Format (NIF) – an RDF/OWL-based format that allows to combine and chain several NLP tools in a flexible, light-weight way
- NLP2RDF – Converting NLP tool output to RDF
- RDFUnit – an RDF Unit-Testing suite
- Smart Data Web – Creation of an industry knowledge base for the German industry.
- TripleCheckMate – Crowdsourcing the evaluation of Linked Data