News
European Conference on Quality in Official Statistics 2026
We had over 20 presentations at the European Conference on Quality in Official Statistics (Q2026), which took place in Šibenik, Croatia, between 3rd and 5th of June 2026.
Special sessions for AIML4OS at Q2026
- Collaboration for Quality: An Overview of the One Stop Shop for AI and ML for Official Statistics project. (WP1) - Slides - Paper
- A New Way to Learn Machine learning and AI: The European AIML4OS Funathon, a Non-competitive Hackathon. (WP6) - Slides
- MLUtils: An Official Statistics Oriented Common Interface for Machine Learning - Slides
- Generalising Earth Observation AI/ML Pipelines for European Statistics (WP7) - Slides
- From Rules to Algorithms: Rethinking Quality Assurance in the Age of AI - Slides
- From the CoP to a Quality Concept for AI/ML in Official Statistics and Back (WP5) - Slides - Paper
- Technical Perspectives on Quality of the Use of AI in Official Statistics (WP5) - Slides - Paper
- The Methodological View: Current Developments of the Total Machine Learning Error Model (WP5) - Slides - Paper
- Landscape of Text Classification with Machine Learning in Official Statistics: Experiences from the AIML4OS Project (WP10) - Slides - Paper
- Automatic Coding in Official Statistics: Enhancing Quality Assurance of the Machine Learning Workflow (WP10) - Slides - Paper
- Measuring Quality of Real and Synthetic Generated Data for Training Statistical Classification Coders (WP10) - Slides - Paper
- Using ML and LLMs for NACE Classification (WP10) - Slides
Speed Talk Session 3, Wednesday, 13:15
- Flexible Machine-Learning-Based Imputation with Options for Sequential Imputation and Predictive Mean Matching - Slides
- Assessing Error in Machine Learning-Based Mass Imputation for Small Area Estimation - Slides
- Standardization of Feature Engineering Implementation Using XML Files in the Imputation Process of Economic Early Estimates - Slides - Paper
- Learning to Fill the Gaps - Slides
All other presentations from AIML4OS at Q2026
- Session 3, Wednesday, 11:00 - Luis Ferreira presented on Unlocking Legacy Statistics with Large Language Models, based on work carried out by WP12 - Slides - Paper
- Session 8, Wednesday, 14:15 - Sónia Quaresma presented on the work of the Ecosystem Monitoring workpackage (WP4), with a talk titled No one is without knowledge except he who asks no questions (Slides , Paper). Klaudia Peszat also presented in this session with a talk on The usefulness of good old ML models in real estate statistics (Slides).
- Session 13, Wednesday, 16:30 - Nina Niederhametner who works with WP10 presented on Using AI for Automatic Data Extraction and Classification in Official Statistics (Slides, Paper). Giulio Massacci also presented in this session with a talk titled: Who Works in the Green Transition? A Patent-Based Analysis of Occupational Change Using Large Language Models (Slides, Paper). The session also includes Luis Sanguiao again, speaking on High frequency time series imputation with auxiliary data: a deep learning approach (Slides).
- Speed Talk Session 10, Thursday, 13:15 Alexandre Cunha from Statistics Portugal (INE) will describe the dataset on inter-firm commercial transactions in Portugal which is used within WP11 for developing a model for reconstructing supply chain networks. Alexandre will discuss the important role of accurate documentation for this highly sensitive dataset. The title of the talk is Whatever Affects one Directly, Affects all Indirectly (Slides).