- Frida Ehrnsten, Eljas Oksanen, Heikki Rantala and Eero Hyvönen: DigiNUMA ja Rahasampo – uusi digitaalinen palvelu rahalöydöistä kiinnostuneille
- Mehwish Alam, Victor de Boer, Enrico Daga, Marieke van Erp, Eero Hyvönen and Albert Meroño-Peñuela: Editorial of Special Issue on Cultural Heritage and Semantic Web
- Heikki Rantala, Eero Hyvönen and Petri Leskinen: Finding and explaining relations in a biographical knowledge graph based on life events: Case BiographySampo
- Heikki Rantala, Eero Hyvönen and Petri Leskinen: Finding relations between entities in a knowledge graph: Case artists of the Getty Union List of Artist Names (ULAN)
|Knowledge Extraction from Natural Language
Case Finnish (and Swedish) Texts
Knowledge Extraction from Finnish Texts
Goals of Research
In Digital Humanities (DH) the data comes in often in textual form (e.g., news, biographies, articles, novels). When publishing such content as Linked Data for DH analysis, the meaning of literal mentions of entities (such as names of persons and places), concepts, relations, events, topics, etc. have to be extracted from unstructured texts and represented as structured semantic data for the computer. In our own work, for example, such knowledge extraction has been needed when developing the Sampo series of semantic portals.
During this research, various natural language processing (NLP) tools and linguistic datasets have been developed. In our view, such tools and resources should be made openly available on the Web as web services. NLP tools and resources would be an important part of the Linked Open Data Infrastructure for Digital Humanities in Finland.
Services for NLP
More information will appear here later.
Dr. Cand. Minna Tamper, Aalto University
Prof. Eero Hyvönen, University of Helsinki (HELDIG) and Aalto