One-Stop-Shop For Artificial Intelligence/Machine Learning for Official Statistics AIML4OS
Project coordinatoor and Co-Partners
Coordinator: Irlande (CSO)
Co-Partners: France (INSEE), Austria (STAT), Denmark (DST), Germany (DESTATIS), Italy (ISTAT),
Luxembourg (STATEC), Norway (SSB), Netherlands (CBS), Poland (GUS), Portugal (INE),
Slovenia (SURS), Spain (INE), Sweden (SCB)
Start date : 01/04/2024 – End date: 31/03/2028
Objective
The use of Artificial Intelligence/Machine Learning (AI/ML) for the production of official statistics is one of the strategic domains that need to be developed further and where coordinated European Statistical System (ESS) action is beneficial. The One-Stop-Shop for Artificial Intelligence/Machine Learning for Official Statistics (AIML4OS) will play an important role in developing innovative solutions with respect to statistical products and processes, allowing for more timely production of official statistics and the delivery of better responses to user needs. The AIML4OS will bring together a consortium of 14 countries (13 Member States and Norway) to develop knowledge and use cases supporting the use of AI/ML-based solutions for the production of official statistics.
Distribution of work
INSEE is coordinating the creation of the ESS AI/ML Lab for the development of AI/ML use cases. It is also contributing to the training of statisticians in these new techniques, in their use for coding in statistical nomenclatures, and in exploring techniques and uses for synthetic data.
The planned activities are divided into 13 work packages with specific objectives, deadlines and results. INSEE is leading work package 3.
- WP1: Project management and coordination
- WP2: Communication and community engagement
- WP3: ESS AI/ML lab: Technical infrastructure and organisational setup
- WP4: AI/ML state-of-play and ecosystem monitoring
- WP5: Standards, methodological and implementation frameworks
- WP6: Knowledge repository and training material
- WP7: Use Case: AI/ML on earth observation data, satellite imagery
- WP8: Use Case: Editing focus - Statistically valid and efficient editing and imputation in official statistics by AI/ML – with a special focus on editing
- WP9: Use Case: Imputation focus - Statistically valid and efficient editing and imputation in official statistics by AI/ML – with a special focus on imputation
- WP10: Use Case: From text to code - Experiences and potential of the use of AI/ML for classifying and coding
- WP11: Use Case: Applying ML for estimating firm-level supply chain networks
- WP12 : Use case: major language models
- WP13 : Use case: synthetic data