Login required to access the wiki. Please register to create your login credentials We apologize for any inconvenience this may cause, but please note that this step is necessary to protect your privacy and ensure a safer browsing experience. Thank you for your cooperation. Direct access: GAMSO , GSBPM , GSIM |
News from the Groups | |||
---|---|---|---|
Blue-skies Thinking | Identifying Topics/Opportunities | CONTINIOUS | Pitches have been made on keeping track of modernisation activities, as well as on ChatGPT, which is turning into a joint endeavour with the ADSaMM group. (There will be a session on ChatGPT for SAS→R code translation on 19th June.) Further pitches have been proposed related to data collection, but otherwise the response to the call for pitches has been slightly modest. |
Non-Probabilistic Surveys | IN PROGRESS | This work is underway, with its own subgroup. | |
Digital Twins | IN PROGRESS | The work of this subgroup has been paused, but the leader is hoping to resume this activity. | |
Open Source Adoption | IN PROGRESS | Kate is making good progress with the themed min-workshops. Additionally, there will be a seminar on open-source at the plenary meeting of the Conference of European Statisticians. | |
Other | |||
Applying Data Science and Modern Methods | Implementing ML-based Solutions in Data Editing | IN PROGRESS | The Data Editing Task Team has now collected a number of use cases on the application and implementation of machine learning methods for data editing and imputation. They have drawn out a list of common issues from these use case and will now work on fleshing out these issues and collating advice for addressing them from the use cases. This advice will provide a set of guidelines for those looking to implement similar projects in the future. |
Understanding and Selecting Models | IN PROGRESS | The Models Task Team are building a taxonomy that provides a high level framework on how to group or understand statistical and data science methods. Given some recent developments the team believe this can be quite a high level set of guidance. In a surprising twist of events the team are testing the usefulness of LLM models as a ‘recommendation engine’ for statisticians and data scientists. They are testing whether different models can provide useful recommendations given a set of defined methodological problems. Early days but seems positive – and might mean that they do not need to create a low level or granular taxonomy. It is likely that they’ll build a set of guidance and examples. | |
Framework for Responsible AI | IN PROGRESS | Progress is ongoing as the Responsible AI team works on the three deliverables: a document that outlines core guiding principles for responsible AI/ML and provides practical guidelines, an assessment tool in the form of a checklist, and a description of a review process (audit) to ensure quality assurance for ML algorithms and AI-based systems. To enhance efficiency and focus, the team members have formed smaller groups, with each group dedicated to a specific pillar or core guiding principle. Additionally, it is worth noting that while attendance has been satisfactory, contributions have been limited over the past few months. | |
Other | At the last two ADSaMM meetings, Christian Ritter (StatCan) presented an MLOps Model – this generated a stimulating conversation and some enthusiasm to continue the discussion subsequently. Several new members have expressed an interest in joining the ADSaMM Group, and so are joining the monthly meetings initially as an ‘observer’. Jointly, ADSaMM & BSTN held two discussions relating to ChatGPT / LLM – both events were highly successful, and will be discussed with the EB. | ||
Capabilities and Communication | Work and Job of the Future | IN PROGRESS | We continue to work on the growth development model. Team did research on the internet and prepared a growth model that was tested by a few countries. General conclusion is that model is generic, intuitive and easy to use. Next steps:
|
Data Analytics | IN PROGRESS | ||
Ethics Management (Data and Business) | IN PROGRESS | Work is focused on the following main points: 1) Analysis of the surveys for the reference book 2) Wiki page was created to collect ideas/information about areas of interest for the reference book and ethical dilemmas . Definitions for the wiki concepts will be sent in the next couple of days. | |
Communication | IN PROGRESS | Working on the document describing communication through the lens of inflation, at the moment putting pieces together from the different countries, like values, objectives, channels and segmentation. Many countries are experiencing issue of communicating inflation regarding disinformation, but also economic education. | |
Ethics Workshop 2023 | NOT STARTED | It was decided to move Workshop to 26-28 March 2024. Organization of workshop on Ethics is at the early stages. Workshop will deal with application of ethics in different areas, like communication, wellbeing, fraud prevention, etc. | |
Other | |||
Supporting Standards | GSIM Revision | IN PROGRESS | Work ongoing as planned, Task Team will soon conclude its work. Clickable and change log for 4 other completed groups (Base, Exchange, Business and Structure) can be found on the Github. Communication paper is currently reviewed by the Task Team. User Guide requires more substantial update due to changes to the model and expected to take a few more weeks to complete. For more details, please check our Github group @ https://github.com/UNECE/GSIMRevision/ |
GSBPM-GAMSO Revision | IN PROGRESS | The Task Team has most recently examined feedback regarding the "Disseminate" phase. Once Disseminate is complete, the group will next consider the "Analyse" phase, and a possible Process phase. For more details, please check our Github group @ https://github.com/UNECE/GSBPM_GAMSO_Revision/ | |
SDMX-DDI-GSBPM | IN PROGRESS | Work ongoing as planned. For more details please check our Google Doc here. | |
Core Ontology for Official Statistics version 2 | NOT STARTED | Work on COOS version 2 is scheduled to start after the summer break. COOS results are available here: https://linked-statistics.github.io/COOS/coos.html | |
Other | Sprint: This was unfortunately cancelled due to unforeseen developments. Call for contributions for: Conference On Smart Metadata for Official Statistics (COSMOS), to be held on 11-12 April 2024 in Paris. The COSMOS conference is a place where the official statistics community can work together to define, share, use, and manage smart metadata. Through sharing of experiences, techniques, and tools, and through forward-looking consideration of the challenges we face as a community, we can better meet the demands of the modern data landscape. The call for contributions can be found here: http://cosmos-conference.org/2024/cfc.html |
- HLG-MOS, 26 Feburary 2023, New York
- Expert Meeting for Machine Learning for Official Statistics, 5-7 June, Geneva
- Expert Meeting on Statistical Data Collection, 12-14 June 2023, online
- Expert Meeting on Statistical Data Confidentiality, 26-28 September, Wiesbaden
- Expert Meeting on Dissemination and Communication of Statistics, 9-11 October, Lisbon
- HLG-MOS workshop on the Modernisation of Official Statistics, 20-22 November 2022, Geneva