Effective and transparent access is what we committed to in this project. A follow-up on the successful CADIAL project, the projects aims to develop management and search infrastructure for the official documentation of the Republic of Croatia using state-of-the-art NLP and machine learning techniques.
TakeLab has joined forces with the Central State Office for the Development of the Digital Society to build and deploy an integrated system for the management of official documentation of the Republic of Croatia. A follow-up on the successful CADIAL project, the project aims to develop the infrastructure and set up the practices for complete digitization and single-point collection of all official documents and allow for a better access to the official documents for all citizens, thus contributing to the efficiency and transparency of governmental institutions.
TakeLab’s objective within this project is to develop a cutting-edge semantic search engine for the official documents of the Republic of Croatia. Under the hood, the engine will feature a number of components, including faceted search, multilabel indexing with EuroVoc descriptors, named entity extraction, and keyphrase extraction, all tailored for the Croatian language. The components will make use of state-of-the-art natural language processing, information retrieval, and machine learning algorithms.
News article about UISUSD available here (in Croatian).
Funder: European Social Fund
Funding: 12,800.000 HRK
Funding scheme: ESF Operational Programme Efficient Human Resources 2014–2020
Grant number: UP.04.1.1.05.0001
Principal investigator (PI): Almir Elezović (project coordinator), Jan Šnajder (FER project leader)
Duration: 3 years (22 January 2018 — 22 January 2021)