Knowledge graphs for societal benefit

Reference number
Coordinator RISE Research Institutes of Sweden AB - RISE SICS AB, Kista
Funding from Vinnova SEK 300 000
Project duration November 2018 - December 2019
Status Ongoing
Venture Personal mobility between societal sectors
Call Funding for staff exchange and artificial intelligence (AI)

Purpose and goal

Authorities collect large amounts of important information that needs to be accessible for both citizens and administrators. Effective and transparent information access is therefore a key issue for Swedish authorities, and is currently managed through traditional methods such as search systems, FAQs and manual customer service. This project investigates how AI can be used to structure information in knowledge graphs, and how these knowledge graphs can be used to build more advanced AI solutions such as intelligent assistants and question-answering systems.

Expected results and effects

The project aims at advancing the collaboration between RISE and Tillväxtverket, with a particular focus on increasing the knowledge transfer regarding the possibilities and limitations of AI. This may facilitate in transforming Tillväxtverket into a more data-driven organization where AI technologies are used practically to improve and streamline make the agency´s operations. For RISE, the project leads to an increased understanding of the actual needs and issues faced by Swedish authorities, and how AI can be applied to increase societal benefits.

Planned approach and implementation

Methods for building knowledge graphs have been a special focus area for the language technology research at RISE SICS in recent years, and this project will be based on the methods developed within this research. More specifically, the project will be based on the use of distributional semantic representations, as well as methods for clustering and categorizing such representations. The resulting knowledge graphs will then be applied in existing collaborative projects on intelligent assistants. The project involves two phases: data collection and knowledge graph generation.

The project description has been provided by the project members themselves and the text has not been looked at by our editors.