DFROG — De Nederlandsche Bank's Machine-Learning-Tested Nowcasting Model for GDP Growth
Netherlands
DNB's DFROG nowcasting model has tracked Dutch GDP growth in real time since 2024, publishing monthly estimates; a decade-long evaluation …
Good practice
★ 87
★ 80
★ 67
★ 67
Mapeamento Regulamentar Automatizado por IA/PLN — o Levantamento do Licenciamento Empresarial da Croácia com Apoio do Banco Mundial
Croatia · Zagreb · See the Croatia profile
Perante a fadiga de reforma causada por inquéritos manuais repetidos, o Ministério da Economia da Croácia e o Banco Mundial testaram um algoritmo de IA/AM/PLN que analisou 110 leis e 1.204 regulamentos, identificando mais de 9.000 potenciais requisitos de licenciamento empresarial — validados em apenas 54%, uma lição honesta sobre os limites do mapeamento regulamentar assistido por IA.
110
Leis analisadas
1,204
Regulamentos analisados
>9,000
Requisitos regulamentares sinalizados
33
Áreas administrativas abrangidas
54%
Validados como relevantes (amostra de artesanato/turismo)
46%
Duplicados ou irrelevantes
Details
- Maturity
- Pilot
- Promoter
- Ministry of Economy, Entrepreneurship and Crafts (MoEEC), Republic of Croatia, with World Bank technical assistance
- Period
- 2019–2020 mapping exercise (report published February 2020); reform programme continued as BER III
- Keywords
- regulatory impact assessment, business licensing, machine learning, NLP, public administration reform
Context
O Ministério da Economia, Empreendedorismo e Artesanato da Croácia, com assistência técnica do Banco Mundial no âmbito do Business Environment Reform Program II, testou em fase piloto um algoritmo de IA/PLN para extrair automaticamente requisitos regulamentares de leis nacionais, em vez de realizar mais um levantamento manual ministerial, entre 2019 e 2020.
Results
O algoritmo processou 110 leis e 1,204 regulamentos, sinalizando mais de 9,000 requisitos potenciais em 33 áreas administrativas; uma revisão manual de validação dos itens de artesanato/turismo concluiu que apenas 54% eram relevantes (13% confirmados como autorizações de atividade e 41% confirmados como condições de inspeção), com 46% de duplicados ou irrelevantes.
Implementation
Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.
Data sources
Where this practice's information was retrieved from, and when.
- C-NAPSE web research — accessed 2026-07-19
- Ministry of Economy, Entrepreneurship and Crafts (MoEEC), Republic of Croatia, with World Bank technical assistance — accessed 2026-07-19
- World Bank mapping report (Feb 2020) — accessed 2026-07-19
- Croatia Week: top-10 business regulation ranking — accessed 2026-07-19
- World Bank: Five Years of Business Environment Reform in Croatia — accessed 2026-07-19
Similar practices you may find useful
Food Inflation Nowcasting with Web-Scraped Prices — Narodowy Bank Polski's Big-Data Model for Real-Time CPI Estimates
Poland
Narodowy Bank Polski scrapes millions of online retail prices to nowcast food inflation ahead of official CPI releases. Peer-reviewed tests …
Payments-Data Nowcasting — the Bank of Canada's Machine-Learning System for Real-Time Macroeconomic Policy Analysis
Canada
Bank of Canada researchers used ML on payments-system data to nowcast GDP and trade during COVID-19, cutting forecast error 20–40% …
Ulysses — Brazil's Chamber of Deputies institutional AI programme for legislative data and staff support
Brazil
Since 2018 Brazil's Chamber of Deputies has run Ulysses, an AI programme that automates legislative-data workflows; a 2025 relaunch made …