Description: Background Better data and statistics are a key enabler for building a better world and addressing the development divide, as well as holding governments to account and improving decision and policy making. In particular, data on key development priorities that are produced with greater frequency, timeliness and granularity are in high demand. In response, National Statistical Offices (NSOs) are increasingly exploring innovative data sources, tools and methods to help address these user needs. The ESCAP Committee on Statistics therefore decided at its 7th session to “feature big data for official statistics in its future work, with an emphasis on sharing country research, experiences and good practices and facilitating capacity development” and to “strengthening legislative provisions and institutional mechanisms to enable national statistical systems to take full advantage of new and innovative technologies while respecting the Fundamental Principles of Official Statistics.” The ESCAP secretariat implements several initiatives to implement these decisions, including the capacity development project entitled the ‘2030 Data Decade - Strengthening the institutional capacity of national statistical offices in Asia and the Pacific to use innovative, new and big data sources for official statistics in support of the 2030 Agenda for Sustainable Development’ (the Big Data Project). The project is funded through the 2030 Agenda for Sustainable Development Sub-Fund of the UN Peace and Development Trust Fund. Through this project, ESCAP is providing technical assistance and related support to countries, as well as developing new knowledge products and facilitating the provision of opportunities for the sharing of achievements among countries in the region. Aims and Objectives The virtual workshop aim to bring together learnings and share experiences in the implementation and application of big data, tools and methods for official statistics from across the Asia-Pacific region. Specific areas of focus will include: Knowledge exchange: Sharing information on implementing and applying big data within official statistics, the challenges encountered and how these were overcome. Discussions will be held with audience participation encouraged. Inspiring others: Demonstrating the art of the possible; showcasing and discussing how alternative data sources can be used within official statistics. Signposting resources: Sharing the new products and tools, as well as the new learning materials and resources as outcomes of the Big Data Project. Establishing further needs: Providing ESCAP with a forum to elicit further learning needs from the Asia & Pacific region. By the end of the virtual regional workshop, participants will have a better understanding of how to overcome the technological, methodological, and human challenges that present when considering the application and implementation of big data in official statistics. Virtual Regional Workshop As the ESCAP Big Data Project draws to a close, the outcomes, challenges and lessons learned from the project workstreams will be showcased and shared across the Asia-Pacific region. A key focus will be on how NSOs have overcome the infrastructural and technological challenges to integrate and apply big data for statistical production. Participants The target audience for the virtual regional workshop is statisticians, methodologists, data analysts and managers within NSOs (or wider National Statistical System agencies) from the Asia-Pacific region. Those who participate in the 11 March session, will be interested in the application of geospatial data and small area estimation within the fields of Environment and Agriculture Statistics, Poverty Statistics, and the Sustainable Development Goals (SDGs); those who participate in the 12 March session will be interested in the use of alternative data sources for calculating Price Statistics. Participants will be invited from across the region’s Official Statistics community and will include those who attended the various Big Data Project workshop, trainings, and High-Level Seminars so that they may learn from others and continue their learning journey. There are no pre-requisites for attending the workshop and there are no limits on the numbers who may attend.
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Machine Learning for Official Statistics: Overview of machine learning techniques and their applications in official statistics, including classification, clustering, and predictive modelling
18 Feb 2025 – 18 Feb 2025
Source: Eurostat (Data extracted on: 01 Dec 2024 )
Description: Webinar 1 - 7 Nov 2024 Webinar 2 - 12 Dec 2024 Symposium - 20-22 Jan 2025 - Dubai, UAE The Data Science Leaders Network (DSLN) and the United Nations Network of Economic Statisticians are organizing a Sprint focused on enhancing NSOs' capabilities in leveraging AI and data science for economic statistics. Through a series of preparatory webinars and an international symposium in Dubai, this event will showcase impactful use cases, explore generative AI applications, and address strategic issues around data privacy, ethical AI, and cross-domain integration. Driven by the goals of the SNA 2025, this Sprint promises to be a pivotal step in empowering the statistical community to harness cutting-edge technologies and shape the future of economic data production and dissemination.
Description: The 12th Statistical Forum of the International Monetary Fund (IMF) will take place in hybrid format (in person and virtually) in Washington, D.C. from November 20 to 21, 2024. The Forum is a platform for policymakers, researchers, the private sector, regulators, and compilers of economic and financial data to come together to discuss cutting edge issues in macroeconomic and financial statistics and to build support for statistical improvements. The theme of this year’s Statistical Forum is Measuring the Implications of Artificial Intelligence (AI) on the Economy. The increased use of AI presents both opportunities and challenges, with significant economic and societal implications. However, the impact of AI on the economy and society remains an evolving field of study. To understand these implications, governments, businesses, individuals require robust and comparable statistics. The 12th Statistical Forum will explore (i) the transformative potential of AI and where its impact will most likely be felt over the short to medium term, (ii) the impact of AI on jobs and productivity, (iii) the distributional implications of AI, (iv) how AI is being used by firms (including statistical agencies) and regulated by governments, and (v) some early attempts to produce official measures of the “AI industry”, “AI investment”, and the “use of AI”. The Forum will provide participants with an opportunity to share experiences and build on topics of mutual interest through presentations and panel discussions.
Description: In this Global Network Webinar, we welcomed Anders Humlum of the Booth School of Business, University of Chicago. He presented on the topic of "The Adoption of ChatGPT." Anders and his co-author Emilie Vestergaard (University of Copenhagen) investigate the adoption of ChatGPT, the icon of Generative AI, using a large-scale survey experiment linked to comprehensive register data in Denmark. Surveying 18,000 workers from 11 exposed occupations, they document that ChatGPT is widespread, but substantial inequalities have emerged. Women are 16 percentage points less likely to have used the tool for work. Furthermore, despite its potential to lift workers with less expertise, users of ChatGPT earned slightly more already before its arrival. Workers see a substantial productivity potential in ChatGPT but are often hindered by employer restrictions and the need for training. Anders covered these and other key findings from the study at this webinar.
Description: The Global Network Webinar on "Unlocking the Power of Large Language Models for Official Statistics: Opportunities, Challenges, and Practical Uses", hosted by the Global Network of Data Officers and Statisticians on 8 October and presented by Amilina Kipkeeva from the UN Economic Commission for Europe (UNECE), provided a comprehensive overview of the growing impact of Large Language Models (LLMs) like ChatGPT and Claude within the official statistics community. The webinar highlighted the significant opportunities LLMs offer to boost productivity and improve user experience, as well as the challenges related to privacy, accuracy, and ethical considerations that require careful attention. The event covered key topics such as the fundamentals of LLMs, their potential applications in various stages of statistical production (from survey design to data dissemination), and the risks and mitigation strategies identified in a recent white paper by the High-Level Group for the Modernisation of Official Statistics (HLG-MOS). Amilina also shared insights from a survey conducted by UNECE on the adoption of generative AI in official statistics, which revealed a broad agreement on the technology's potential impact, as well as the need for formal policies and staff training to address concerns around data security, privacy, and ethical use. The webinar highlighted several compelling use cases, including the Bank of International Settlements' AI-powered metadata editing tools, the Australian Bureau of Statistics' successful application of LLMs for occupation code updates, and the efforts of Statistics Canada and Deutsche Bundesbank in leveraging LLMs for report generation and content creation. The presentation emphasized the importance of international collaboration and knowledge sharing as the official statistics community navigates this rapidly evolving landscape. During the Q&A session, participants raised questions about the development of policies and governance frameworks to guide the responsible use of LLMs, as well as the potential for these technologies to address challenges around human resource gaps in statistical organizations, particularly in developing countries. The discussion also touched on the distinction between internal and external use of LLMs, and the importance of ensuring the accuracy and reliability of LLM-generated outputs when engaging with the public.
Description: This presentation scrutinises the transformative potential of Large Language Models (LLMs) in survey research, focusing on three critical areas: questionnaire design, synthetic data creation, and the role of LLMs as qualitative interviewers. In the domain of questionnaire design, the lecture delves into if and how LLMs can construct contextually accurate and highly effective survey items. However, there are valid concerns about the model’s understanding and potential biases, which we will critically evaluate. She also discusses LLMs’ ability to fabricate synthetic data, preserving core statistical properties whilst ensuring privacy. Here too, the ethical implications and the potential for misuse of this capability pose challenges that need to be addressed. Lastly, the lecture explores how LLMs, with their human-like conversational ability, can act as qualitative interviewers, allowing in-depth information gathering at scale. Yet, questions about their ability to fully capture the complexity and subtleties of human interaction and response also remain. The underlying theme of this talk is the question on how research in this space should be structured.
Title in Arabic: منهجية علم البيانات في الاحصاء الرسمي
Organizer(s): ESCWA AITRS
Description: تُعدّ البيانات عنصرًا هامًا لاتخاذ القرارات ورسم السياسات السليمة ويلعب علم البيانات دورًا رئيسيًا في تحليل هذه البيانات واستخلاص المعلومات منها، بينما يُساهم الإحصاء الرسمي في ضمان جودة البيانات وسلامتها. وتهدف هذه الدورة التدريبية عن بعد إلى تعريف المشاركين بمفاهيم منهجية علم البيانات في الإحصاء الرسمي، وتزويدهم بالمهارات اللازمة لتطبيقها في مختلف مجالات العمل الاحصائي. تأتي هذه الورشة في سياق الاهتمام الإقليمي والدولي بتطوير القدرات الإحصائية وذلك من خلال المقترح الذي أيدته لجنة الإحصاءات التابعة للأمم المتحدة المعد من اللجنة الاستشارية للخبراء عن البيانات الضخمة وعلوم البيانات للإحصاءات الرسمية لإدماج البيانات الضخمة وعلوم البيانات في العمل اليومي لمكاتب الإحصاء الوطنية، وإنشاء شبكة من قادة علوم البيانات في هذه المكاتب. بالإضافة إلى ذلك، تأتي هذه الورشة بعد ورشة عمل إقليمية ناجحة حول موضوع البيانات الضخمة في الإحصاء الرسمي، التي نظمتها الاسكوا بالتعاون مع المعهد العربي للتدريب والبحوث الاحصائية. تهدف الورشة بالأساس الى تعزيز قدرات المشاركين في مجال الإحصاء الرسمي باستخدام أساليب علم البيانات وذلك من خلال: * تعريف المشاركين بمفاهيم أساسية في علم البيانات والإحصاء الرسمي. * شرح منهجية علم البيانات في الإحصاء الرسمي. * إكساب المشاركين مهارات تحليل البيانات واستخلاص المعلومات منها. * تمكين المشاركين من تطبيق مهارات علم البيانات في الإحصاء الرسمي من خلال حالات احصائية عملية.
Description: This Joint Webinar of the Intersecretariat Working Group on Household Surveys (ISWGHS) and of the Global Network of Data Officers and Statisticians, which was attended by around 210 participants, featured Stephanie Eckman from the Social Data Science Center of the University of Maryland. She discussed how insights from survey methodology can help improve the quality of training and feedback data used for machine learning models like large language models (LLMs) which can improve model performance. Stephanie highlighted two key areas where survey methodology principles are relevant: 1) Representation - ensuring the labelers who train the models are representative of the target population, and 2) Measurement - ensuring the labeling instruments and procedures yield high-quality labels. On representation, she showed how characteristics of the labelers can impact the labels and resulting models, creating selection biases. If many labelers are from specific countries/regions then this can introduce demographic biases. One possible solution to adjust for this is using statistical weights for the different groups of labelers. On measurement, she described how aspects of the labeling interface like question wording, order, and format can significantly impact the quality of labels, which then impacts the trained model performance. Through an experiment, she demonstrated the importance of carefully designing labeling instruments akin to survey design principles. The Q&A covered topics like using sentiment analysis instead of binary labels, accounting for survey fatigue, leveraging existing survey data for model training, and concerns around cultural biases embedded in LLMs trained primarily on data from western, rich countries. Overall, Stephanie made a strong case for interdisciplinary collaboration between the survey research and AI/ML communities to improve training data quality and develop more accurate, unbiased AI models aligned with human values.
Organizer(s): UN Global Pulse UNFPA WFP WHO Finland
Description: Three Tales of Transformation: Catalysing innovations for the 2030 Agenda with the UN Global Pulse Scale Accelerator Organized by: UN Global Pulse, the Government of Finland, United Nations Population Fund, World Food Programme and World Health Organization Join teams from WFP, WHO and UNFPA as they share their journeys of catalysing their promising innovations to turn […]
Description: The members of the Data Science Leader's Network (DSLN), under the umbrella of the Committee of Experts on Big Data and Data Science for Official Statistics, have agreed to develop a DSLN Playbook with the objective to provide a comprehensive guide for integrating data science into the work of National Statistical Offices, and to offer a systematic approach to piloting projects, pooling resources, documenting successes, and overcoming institutional barriers. The "playbook" format has been chosen as a way of capturing (in a user-friendly manner) a structured set of practical guidelines designed to help NSOs achieve specific objectives towards the mainstreaming of data science in the day-to-day business of official statistics, providing both strategic and tactical guidance. It aims to be a practical, hands-on resource that can be easily consulted and followed by practitioners at the frontlines of statistical work, with step-by-step procedures, best practices, checklists, and real-world examples that help a team understand not just the "what" but also the "how" of implementation. The event will be an opportunity to present and discuss the progress towards the development of the Data Science Playbook, based on the outcome of the 2nd Sprint of the Data Science Leaders' Network, which took place in Dubai from 22 to 24 January, as part of the International Seminar on Data Science for the Statistical and Transport Communities.
Source: Eurostat (Data extracted on: 24 Jan 2024 )
[+] More
Organizer(s): Eurostat Icon-Institut
Description: Main objectives of the course are: Introducing the participants to the concept of Big Data, the associated challenges and opportunities, and the statistical methods and IT tools needed to make the use of Big Data effective in official statistics. Overviewing statistical methods and IT tools available for Big Data usage in Official Statistics
Target Audience: Official statisticians (including managers) to be involved in big data activities and having no specific knowledge on this subject; Official statisticians (including managers) who, without being directly involved in big data activities, need basic knowledge on the use of big data in official statistics.
Description: The webinar is an overview of the report on the 10-years of use of Big Data and Data Science for Official Statistics. The survey details will be described along with results and recommendation based on the survey. The full report of the 10-year review has been released as a background document to the Statistical Commission and is available here.
Description: This course is presented by the ISI Statistical Capacity Building (SCB) Committee. It is available for free to everyone. The course includes an introduction to (descriptive) statistics, and modules on sampling, probability, statistical inference, experimental design, categorical data, non-parametric methods, and linear regression.
Source: ESCAP SIAP (Data extracted on: 23 Jan 2024 )
[+] More
Organizer(s): ESCAP SIAP ESCAP ADB
Description: This 8-week course developed by SIAP in partnership with the Asian Development Bank (ADB) introduces machine learning as a tool for using either traditional (surveysmicro data…) or non-traditional (big data) data sources to produce high quality predictions for Official Statistics or Sustainable Development Goals (SDGs) indicators. It provides an opportunity for participants to explore and manipulate the techniques of Machine Learning and their links with traditional statistical methods. The 6 modules (+1 module with recalls/prerequisites) aim at providing an overview of the current methods and applications of Machine Learningthrough simplified theoretical conceptspedagogical case studies and interactive resources.The course is not basednor does it requirea particular software. Howeverreproducible examples on either simulated or real data are provided using the R/RStudio environment. Some Python procedures and packages are also provided. The weekly webinars are planned on Wednesdays at 4 p.m. Japan time.
Organizer(s): UNSD Global Partnership for Sustainable Development Data INE Uruguay UNCEBD UN MGCY AWS UN Big Data Regional Hub in Brazil UN Big Data Regional Hub in China UN Big Data Regional Hub in Rwanda UN Big Data Regional Hub in the United Arab Emirates ECLAC Statistics Indonesia Statistics Canada ONS United Kingdom Oblivious OpenMined World Food Forum
Description: Leverage technology and your expertise to develop innovative solutions for a more sustainable and equitable future. Join us on this journey of innovation, exploration, and problem-solving as we harness the power of data to create a more sustainable and resilient world. Datathon participants will develop innovative data-driven applications, tools or statistical models combining geospatial data with other data sources to help advance the implementation of the Sustainable Development Goals. Register today!
Description: Leading up to the UN Datathon 2023 UNSD in cooperation with the UN Committee of Experts on Big Data and Data Science (UNCEBD) and with support of the UN Committee of Experts on Geospatial Information Management (UN-GGIM) organized webinar exploring the crucial synergy between geospatial data and statistics in driving progress towards the Sustainable Development Goals (SDGs) within the realm of big data.Participants were invited to discover why integration matters, how to unlock creative data visualization techniques, and how to gain a deeper understanding of utilizing geospatial insights for the SDGs. The webinar, which was attended by approximately 480 participants, had a lineup of three excellent speakers in the field of integration of geospatial information and statistics. The first speaker was Claudio Stenner, co-chair of the United Nations Expert Group on the Integration of Statistical and Geospatial Information. He is a geographer with experience in regionalization, urban geography and geoprocessing and works at the in Brazilian Institute of Geography and Statistics (IBGE) where he is in charge of Brazil's 2022 Demographic Census on issues related to geography and its integration with statistics. In his talk he demonstrated the importance and potential advantages for society in the integration of statistical and geospatial information and introduced the attendees to the Global Statistical Geospatial Framework (GSGF) which allows this integration. The second speaker was Ken Field, who after 20 years of academic career, now talks and writes about cartography, teaches, and makes maps at Esri in California. He teaches a Massive Open Online Course on Cartography which had so far over 200,000 participants and is the author of two award-winning books, Cartography (2018) and Thematic Mapping (2021). In his talk Ken said that every map is a product of its maker, its reader, and multiple competing contexts. He argued that maps are rarely right or wrong but, instead, offer different versions of the truth. His talk explored how the design of thematic maps mediates the message using a range of examples based on Covid-19 and election maps. The ideas and techniques explored during his talk provide a strong foundation for use on any other empirical data, big or small, and including the SDGs. The third speaker was Britta Ricker who is a Digital Geographer and faculty member within the Copernicus Institute of Sustainable Development at Utrecht University in the Netherlands. Through her teaching and research, she focuses on the use of open data, open software, and open science for cartographic production. Her work bridges qualitative and quantitative research by identifying practical solutions for spatial data collection, management and effective geovisualization of the UN Sustainable Development Goals at the local and global level. In her talk she shared examples from her open access book published together with the Sustainable World” as well as some ideas for future SDG spatial data solutions. The webinar is part of a series organized by UNSD to prepare the participants of the upcoming UN Datathon 2023 for this big data and data science competition. The UN Datathon will take place in November world-wide and at 8 in-person locations with the main venue in Montevideo, Uruguay.
Description: For this Global Network Webinar we were happy to have with us Pau Garcia, media designer and founder of Domestic Data Streamers, who talked about Artificial Ignorance and Probabilistic Stories. This talk discussed the ethical considerations of utilizing AI in creative industries, exploring its power to produce both astonishing potential and dangerous consequences. It will examine how AI can have unseen impacts and the responsibility of creatives to ensure the safety of their work and the people they work with. Pau Garcia a media designer and founder of Domestic Data Streamers. Since 2013, the Barcelona-based studio has been producing immersive “info-experiences” for institutions like the United Nations, Tate Modern, and Citizen Lab. Garcia is chair of the Master in Data in Design at ELISAVA and has lectured at the Hong Kong Design Institute, the Royal College of Arts, Politecnico di Milano, and the Barcelona School of Economics. He also founded HeyHuman!, an artist residency program that combines music, journalism, data for art research, and social justice.
Description: The UN Datathon (3-6 November) welcomes participants from around the globe of all ages and backgrounds, as long as you have a basic level of data science experience. If you are passionate about data-driven solutions at local and global levels, and eager to contribute to the implementation of the UN Sustainable Development Goals, this event is for you!
Description: During the webinar we will announce the survey on the 10-year review of the UNCEBD. The objective of the webinar is to raise awareness for the survey and generate interest by statistical agencies to complete the survey. We would like to take this opportunity to evaluate, if NSOs and other statistical agencies are familiar with the work of the UN Committee of Experts on Big Data and Data Science for Official Statistics (UNCEBD, previously also known as the Global Working Group on Big Data for Official Statistics).
Description: For this Joint FAO and Global Network Webinar we had with us Carola Fabi, Marco Scarnò and Craig Steforn Matadeen from FAO who presented on Essence, an integrated framework for documents retrieving and analysis developed by FAO. “Essence” stands for Expert Search Semantic ENriChmEnt and is a web tool developed by the FAO Data Lab that automatizes searching for scientific articles and identifies the most relevant ones through the use of artificial intelligence (AI). The AI method learns and extends users’ previous selections by recognizing patterns in their texts. Essence offers a semantic search engine and interactive filters to navigate the stored data, plus data visualization functions or procedures to extract statistical values from unstructured texts.
Description: From 8 March to 10 May 2023, FAO's Office of Chief Statistician organized a Webinar Series on Earth observation data for agricultural statistics. The webinar series raised awareness of the EOSTAT project and highlighted FAO's work in building countries' capacity on the use of Earth observation data for the production of agricultural statistics. The EOSTAT webinar series also discussed the main innovations introduced in the field of EO based crop statistics implemented by FAO through R&D and testing in countries in collaboration with the academia. These include: 1) Standardization of EO methods to produce annual land cover/use maps 2) Crop type mapping in the context of in-situ data scarcity 3) Coupling of EO data with physical based crop yield modelling Moreover, the series shed light on how the innovations brought by the EOSTAT offer a solution to overcome the challenges connected with the collection and predication of crop statistics (acreage and yield) using both traditional field survey-based methods and the EO based methods. Finally, the webinar series provided a platform to enhance collaboration and potentially mobilize resources. The FAO Webinar Series "Earth observation data for agricultural statistics" consisted of six webinar sessions. Simultaneous interpretation was provided in English and Spanish. Session 1 (8 March 2023, 15:30 – 17:00) Recording (Passcode: =Y9v=5WF) Presentation (Pietro Gennari) Presentation (Lorenzo de Simone) EOSTAT project overall presentation Organized jointly with the Global Network of Data Officers and Statisticians Speakers: Pietro Gennari, FAO & Lorenzo De Simone, FAO Session 2 (20 March 2023, 15:30 – 17:00) Recording (Passcode: 4L1vtay%) Presentation (Lorenzo de Simone) Presentation (Prof. Bruno Basso) Crop yield mapping and yield statistics Speaker: Lorenzo De Simone Guest: Prof. Bruno Basso, Michigan State University Session 3 (4 April 2023, 15:30 – 17:00) Recording (Passcode: f*v2Z^!n) Presentation (Sophie Bontemps) Crop type mapping and acreage Speaker: Lorenzo De Simone, FAO Guest: Sophie Bontemps, Université of Louvain Session 4 (13 April 2023, 15:30 – 17:00) Recording (Passcode: A2Y7&&Uu) Presentation (William Ouellette) Standardized land cover classification for land cover statistics Speaker: Lorenzo De Simone, FAO Guest: William Ouellette, CEO at SoilWatch Session 5 (27 April 2023, 15:30 – 17:00) Recording (Passcode: 4@c#56=D) Presentation (Sophie Bontemps) EO augmented survey design, in-situ data standards, and best practices in georeferencing Speaker: Lorenzo De Simone, FAO Guest: Sophie Bontemps, Université of Louvain Session 6 (10 May 2023, 15:30 – 17:00) Recording(Passcode: kf7Nh?QK) Presentation (Sherrie Wang) Crop field boundaries mapping using machine learning and very high-resolution data Speaker: Lorenzo De Simone, FAO Guests: Sherrie Wang, MIT Lisa Rebelo, Digital Earth Africa For more information, please visit the FAO-EOSTAT project page.
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Description: Click here to go to the event The African Centre of Statistics (ACS) of the United Nations Economic Commission for Africa (UNECA) is launching the first of StatsTalk-Africa webinar series. The series will provide a space for dialogues about data, statistics, and innovative tools with data experts and users within ECA, Africa and rest of the world. The first of the monthly webinar series will focus on: Demystifying Big Data and Official Statistics in Africa and will take place on Friday, 28th April 2023 from 12:30 to 13:30 EAT via Microsoft Teams. The webinar will explain the link between Big Data and Official Statistics and how they can be used to inform better planning, policymaking towards achievement of Agenda 2030 and Agenda 2063, the Africa We Want. The event is expected to increase data literacy, understanding of the big data situation, needs and uses in Africa; and share innovative tools and methods for data production and management. This webinar specifically aims to: Serve as a knowledge-sharing and exchange platform. Demystify and promote greater understanding of key statistical concepts and alternative data sources that could be harnessed in the African context. Register in advance for this meeting here. Microsoft Teams meeting Join on your computer, mobile app or room device. Click here to join the meeting Meeting ID: 395 536 119 290 Passcode: Yuv6wN Download Teams | Join on the web Join with a video conferencing device. unitevc@m.webex.com Video Conference ID: 124 638 313 6 Alternate VTC instructions
Description: Background In 2022, ESCAP developed a pilot project specifically targeted at supporting NSOs in Asia and the Pacific to address this complexity. The project aims to work with NSOs in the region to pilot the use of big data for producing a sample of the 46 environment-gender indicators. Cambodia and Mongolia submitted their expressions of interest in participating in the pilot to produce one indicator per country. For Cambodia, the indicator is the proportion of population with access to electricity by sex, while Mongolia focuses on measuring the proportion of population in Ulaanbaatar living in Gers by sex. More specifically, the project team utilized Google Earth Engine to derive nighttime light data for all communes in Cambodia and combined the results with average household size and female ratio to estimate the indicator. In the case of Mongolia, the team designed an object-based image analysis model on QGIS using the Orfeo Toolbox and a random forest classifier to detect Gers. In both cases, the team conducted sensitivity analyses and compared the results against indicators produced from censuses and/or surveys. The two pilots produced promising results and lessons learned. The project team recently conducted an Expert Group Meeting to present findings, lessons learned and challenges from implementing the pilots to experts in the field and seek their feedback on how the models can be improved. Insights from the meeting has been integrated into the guidance documents which record processes throughout the entire pilot and recommend next steps for the two NSOs to improve their usage of big data in official statistics production. This Stats Cafe session , Stats Café Home: Upcoming events Concluded events in 2022 2021 events 2020 events
Description: Traditional surveys are not well-equipped to measure certain concepts of interest such as expenditures, time use or travel behavior due to high burden placed on participants. Facts or behaviors that are difficult to measure through self-report can be measured using new technologies: smartphone apps, sensors, and wearables. For example, accelerometers in smartphones and fitness bracelets can objectively measure physical activity, screen time apps can measure (social) media use. Another possibility is to augment surveys with administrative data or data from digital platforms such as Google, Youtube, Instagram that participants can provide to researchers through data donation, or consent to data linkage. However, to ensure representation, participants have to be willing and able to use their devices to perform such tasks. If participants differ from nonparticipants in key outcomes, research results can be biased. In this webinar, I will present the results of several randomized experiments on the mechanisms of willingness and consent to collect data using smartphone sensors, apps, and wearables in general population surveys, and the extent of nonparticipation bias assessed by linkage of survey data to sensor and administrative data. I will further focus on how these mechanisms translate to data donation of digital trace data, what opportunities and challenges such novel data collection methods hold for the social sciences and official statistics, and outline future research agenda.
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Source: Eurostat (Data extracted on: 19 May 2023 )
[+] More
Organizer(s): European Court of Auditors
Description: This event will explore the future of official statistics in achieving independence and accountability in the age of big data. High-quality official statistics are a critical part of evidence-based decision making by governments, businesses, researchers and citizens. They can also promote transparency and accountability. Following the publication of our special report on the quality of European statistics in November 2022, the European Court of Auditors (ECA) will hold an online conference on Tuesday 28 March 2023 to discuss the future of official statistics from the perspective of independence and accountability. We will also discuss how the statistical landscape will be reshaped by using new data sources such as big data that could lead to a paradigm shift in data collection and analysis. The conference will feature two panel discussions. Following the keynote address by Professor Enrico Giovannini, former chair of the European Statistical Governance Advisory Board (ESGAB), the first panel will focus on the future of governance, independence and accountability in official statistics. The second panel will move towards the future of statistical product development following the keynote address by Dr Sallie Ann Keller from the US Census Bureau. ECA reporting Member for the special report on European statistics, Ildikó Gáll-Pelcz, will deliver the opening address, which will be followed by a short presentation by Athanasios Koustoulidis, the head of task for the audit. In addition to the two keynote speakers, six senior executives from international organisations (Eurostat and the IMF) and national statistical offices (Sweden and Poland) will participate in the two panels.
Description: The Expert Group Meeting Big Data in Official Statistics in Asia and the Pacific will be held on 23 March 2023, 14:00–16:00 hours (Bangkok time). Register here: https://forms.office.com/e/3ejwKuu0CE Join the meeting via Teams Concept note and Agenda Aims of the workshop The purpose of this Expert Group Meeting is to present findings, lessons learned and challenges from implementing the pilots to experts in the field and seek their feedback on how the models can be improved. The project team and participants can learn from the experiences and comments of experts in the field of earth observation data and nighttime light data for statistics production. Insights from the meeting will be collated and integrated into the guidance document which records processes throughout the entire pilot and recommends next steps for the two NSOs to improve their usage of big data in official statistics production. , HOME - Project: measure the nexus between environment and gender Project countries: Cambodia Mongolia Resources Regional Events
Title in Spanish: Serie de seminarios web de la FAO: Datos de observación de la Tierra para estadísticas agrícolas SESIÓN 2: Mapeo de rendimiento de cultivos y estadísticas de rendimiento
Organizer(s): FAO
Description: El lunes 20 de marzo, nuestro invitado especial, el Dr. Bruno Basso, profesor del Departamento de Ciencias Ambientales y de la Tierra de la Michigan State University, hablará sobre el mapeo del rendimiento de cultivos y las estadísticas de rendimiento. El evento se realizará en inglés con interpretación simultánea a español y podrá seguirse exclusivamente de forma virtual. Enlace de registro: https://fao.zoom.us/webinar/register/WN_3i2xcYuYS8qGEyKojo2-WA El propósito de esta serie de seminarios web es dar a conocer el proyecto EOSTAT y destacar el trabajo de la FAO en la creación de capacidad de los países en el uso de datos de observación de la Tierra para la producción de estadísticas agrícolas, contribuyendo así a la modernización de los sistemas y protocolos para las estadísticas nacionales. También discutirá las principales innovaciones introducidas en el campo de las estadísticas de cultivos basadas en observaciones de la tierra implementadas por la FAO a través de I+D y pruebas en países en colaboración con la academia. Éstas incluyen: 1) Estandarización de los métodos de observaciones de la tierra para producir mapas anuales de cobertura/uso de la tierra 2) Mapeo de tipos de cultivo en el contexto de escasez de datos in situ 3) Acoplamiento de datos de observaciones de la tierra con modelos físicos de rendimiento de cultivos Además, la serie arrojará luz sobre cómo las innovaciones aportadas por EOSTAT ofrecen una solución para superar los desafíos relacionados con la recopilación y pronóstico de estadísticas de cultivos (superficie en acres y rendimiento) utilizando métodos tradicionales basados en encuestas de campo y métodos basados en observaciones de la tierra. Finalmente, la serie de seminarios web proporcionará una plataforma para mejorar la colaboración y potencialmente movilizar recursos. El lunes 20 de marzo, nuestro invitado especial, el Dr. Bruno Basso, profesor del Departamento de Ciencias Ambientales y de la Tierra de la Michigan State University, hablará sobre el mapeo del rendimiento de cultivos y las estadísticas de rendimiento. El evento se realizará en inglés con interpretación simultánea a español y podrá seguirse exclusivamente de forma virtual. Para obtener más información sobre los seminarios web pasados y próximos planificados, visite la página dedicada aquí. Acerca de nuestro invitado: Bruno Basso es Profesor Distinguido John A. Hannah y Profesor de la Fundación MSU de Ciencias Ambientales y de la Tierra en la Michigan State University. Es científico de agroecosistemas y modelador de sistemas de cultivos con interés en la sostenibilidad a largo plazo de los sistemas agrícolas, agricultura digital, bioeconomía circular. Su investigación se centra en evaluar y modelar la variabilidad espacial y temporal del rendimiento de los cultivos, el carbono orgánico del suelo, las emisiones de GEI, el agua y los flujos de nutrientes en los paisajes agrícolas en los climas actuales y futuros. Posee patentes globales sobre IA, sensores remotos y sistemas de modelos de cultivos para evaluar la productividad de las tierras de cultivo y la sostenibilidad ambiental. Para obtener más información, visite la página del proyecto FAO-EOSTAT y el sitio web de estadísticas de la FAO.
Title in Spanish: Demostración del uso datos de observación de la tierra en el desarrollo de estadísticas agrícolas “El caso de Ecuador"
Organizer(s): FAO
Description: Esta demostración tiene como objetivo proveer a los funcionarios gubernamentales responsables de la producción y análisis de información de estadísticas agrícolas, un espacio abierto para la solución de dudas sobre el uso de datos de observación de la tierra en la modernización y producción de estadísticas agrícolas. La FAO con el ánimo de desarrollar capacidad en los países en el uso de datos de observación de la tierra para la producción de estadísticas agrícolas ha venido promoviendo diferentes actividades en las cuales los países son actores fundamentales, este es el caso de Ecuador, quién ha trabajado fuertemente en mejorar y modernizar su sistema estadístico agrícola. En línea con este trabajo, la FAO busca apoyar ahora a más países de la región en el desarrollo y mejoramiento de estas capacidades en países como Chile, Colombia y Perú, y para ello, realizará una demostración de los resultados obtenidos en Ecuador. Expositor: Lorenzo De Simone, PhD.Technical Adviser Geospatial, Office of the Chief Statistician, FAO. Asesor de la FAO sobre el uso de Observaciones de la Tierra para el monitoreo de la Agricultura y los Recursos Naturales. También es experto en ciencia de datos, sistemas de información geográfica y teledetección (PhD), cuenta con conocimiento de geoestadística y ciencia ambiental, desarrollo de TI y gestión de calidad. Grabación (Código de acceso: x.Bh!Sf3)
Description: During this Joint Global Network and FAO Webinar on Earth Observation Data for Agricultural Statistics Lorenzo De Simone and Pietro Gennari provided us with an overview of the EOSTAT project. This webinar was the first in a Webinar Series on Earth Observation (EO) Data for Agricultural Statistics organized by the Food and Agriculture Organization (FAO) from March to May which will raise awareness of the EOSTAT project and will highlight FAO's work in building countries' capacity on the use of earth observation data for the production of agricultural statistics. The speakers gave some high-level insights of the main innovations introduced in the field of EO-based crop statistics implemented by FAO through R&D and testing in countries in collaboration with the academia, including the 1) Standardization of EO methods to produce annual land cover/use maps; 2) Crop type mapping in the context of in-situ data scarcity; and 3) Coupling of EO data with physical based crop yield modelling. Pietro and Lorenzo gave an overview of how the innovations brought by the EOSTAT offer a solution to overcome the challenges connected with the collection and predication of crop statistics (acreage and yield) using both traditional field survey-based methods and the EO-based methods. They also spoke about how the FAO webinar series will provide a platform to enhance collaboration and potentially mobilize resources.
Description: The United Nations Committee of Experts on Big Data and Data Science for Official Statistics (UNCEBD), Major Group for Children and Youth (MGCY) and Statistics Indonesia (BPS) hosted the "2022 UN Big Data Hackathon" in November 2022. It is jointly organized by the United Nations Statistics Division (UNSD), Global Platform Regional Hubs (Rwanda, UAE, Brazil & China), UN Global Pulse, Asian Development Bank (ADB), Islamic Development Bank (IsDB), UK ONS Data Science Campus, Statistics Canada, and United Nations Conference on Trade and Development (UNCTAD), in consultation with the members of the Task Teams of UN Committee of Experts on Big Data and Data Science. Join us for the Closing Ceremony where we will be announcing and celebrating the winners of the Hackathon.
Description: In recent years, almost every government has been faced with very serious challenges, such as the global health pandemic, supply chain disruption, rising energy and food prices, and decreasing household budgets. To handle these crises in the right way, our leaders need the right data at the right time. National statistical offices are tasked to provide these trusted, relevant, timely and high-quality data. All of those data are very sensitive in terms of private information on persons or businesses. To gain access to the sensitive data while guaranteeing that privacy will be preserved, privacy-enhancing technologies (PETs) are receiving increased attention. Whereas legal arrangements on data sharing can lead to unwanted breaches, the promise of PETs is that privacy is guaranteed. If you cannot see the original data at any time, you cannot by accident reveal any original information. In this webinar, experts of the task team on PETs will launch the "UN Guide on PETs for Official Statistics". The program is as follows.
Source: Eurostat (Data extracted on: 03 Feb 2023 )
[+] More
Organizer(s): Eurostat Icon-Institut
Description: Main objectives of the course are: * Introducing the participants to the concept of Big Data, the associated challenges and opportunities, and the statistical methods and IT tools needed to make the use of Big Data effective in official statistics. * Overviewing statistical methods and IT tools available for Big Data usage in Official Statistics.
Target Audience: Official statisticians (including managers) to be involved in big data activities and having no specific knowledge on this subject; Official statisticians (including managers) who, without being directly involved in big data activities, need basic knowledge on the use of big data in official statistics.
Description: On 2 December, the Statistics Technical Network will organize a webinar on the fundamentals of Deep Learning, where FAO colleagues from the Statistics Division will present the benefits of fast.ai. WHEN | Friday, 2 December 2022, 14:00-15:00 Rome time | REGISTER HERE Nowadays neural networks are widely used in a wide range of applications. In some areas, such as Computer Vision (CV) and Natural Language Processing (NLP), this new technology has outperformed previous state-of-the-art models. The FAO Data Lab is using neural networks to tackle real-world problems on daily data (such as classification of tweets and images) by exploiting fast.ai, a Python library built on top of PyTorch. fast.ai allows to build neural nets from scratch, tailored to the project's purposes, in a few hours with a perfect combination of model customization and speed (to create, train and deploy models). In this webinar, we will explore the potential of this library showing a couple of case scenarios, in the field of NLP and CV, that can be solved with fast.ai. The aim of this presentation is to share our approach to any potential user and provide suggestions about solving tasks that could be problematic with other standard methods by using this off-the-shelf and powerful tool. Introduction Carola Fabi, Senior Statistician, FAO Statistics Division Presenter Gianfausto Bottini, Data Scientist, FAO Data Lab For more information about the FAO Data Lab, please visit: https://www.fao.org/datalab/website/web/home
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Description: While the BigData-native statistical community is growing larger, sampling statisticians seem to grow divided between enthusiastic and worried. Is BigData also a big step ahead to extract trustful information and actual knowledge from data? Is BigData underplaying sampling theory? Supplanting it as a low-cost futuristic option? In this webinar I shall try and decipher the multifaceted relationship connecting BigData and Sampling methodology, starting with the blurry definition of BigData, discussing the non-probabilistic data generating mechanism, passing through different kind of data, of application contexts and goals, to end with some very personal considerations and views.
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Description: Concept National Statistics Offices (NSOs) are facing ever-growing demands for timely and quality official data and statistics due to rapid technological developments and COVID-19. To meet these demands, NSOs are exploring other non-traditional sources of data, but at the same time, they need to maintain data quality, ethics, and privacy protection standards that align with the UN Fundamental Principles of Official Statistics. In pursuing this goal, they face several new challenges including building partnerships with private data holders, investing in new infrastructures and human resources, and updating data governance arrangements. The UN Statistical Commission recognized the needs to address these challenges and established the UN Committee of Experts on Big Data and Data Science for Official Statistics (UN-CEBD) at its 45th session to facilitate coordinated efforts and explore innovative solutions. UN-CEBD has eight Tasks Teams including the Task Team on Big Data which is responsible for investigating the benefits and challenges of using big data for monitoring and reporting on SDGs. UN-CEBD arranges annual International Conferences on Big Data and Data Science for Official Statistics and the latest Conference was held in Yogyakarta from 7th to 11th November 2022 with the theme of Global Challenges and the Importance of Relevant and Timely Data. Workshops on Earth Observation Data, AIS, Cell Phone Data, Data Scanner, Machine-Learning and Privacy-Preserving Techniques and UN Big Data and UN PET Lab hackathons also accompanied the main Conference. The Conference generated insightful discussions, and this ESCAP Stats Café aimed to share the key takeaways and ways forward with relevant stakeholders in Asia and the Pacific region. The session started with opening remarks from Ms Rachael Beaven—Director of Statistics Division, ESCAP—and Mr Ronald Jansen—Assistant Director of UN Statistics Division, DESA, UNSD. Following the showcase, a video highlights activities from the Conference. Speakers from NSOs and UNFCCC who attended the Conference and UN Big Data Hackathon also shared their experiences, lessons learned, and inputs on ways forward. Lastly, the ESCAP’s big data guide on "Using big data for official statistics: Key considerations when using mobile phone data" was disseminated. , Stats Café Home: Upcoming events Concluded events
Machine Learning for Official Statistics
and SDG Indicators
21 Nov 2022 – 15 Jan 2023
Source: ESCAP SIAP (Data extracted on: 31 Oct 2022 )
[+] More
Organizer(s): ESCAP SIAP ADB
Description: The course is designed for personnel working in the field of statisticswhose main responsibilities include data analysis of SDG indicators and related statistics with a specific target on data scientists from NSOs with an experience in both statistical modelling (regression analysispredictionclassification...) and with programming or algorithmic skills. Although no programming will be required to follow and succeed in the coursethe pedagogical materials include R codein the form of reproducible markdown notebooksas well as some Python resources and code.
Description: The first training aimed to provide an introduction in using big data in official statistics. It is a part of the training series under the 'Using Big Data to Measure the Nexus between Environment and Gender in Asia and the Pacific" project to support the beneficiary countries to utilize big data for piloting the production of a sample of environment-gender indicators to assist the countries in improving the availability, inclusiveness, and sustainability of quality data and official statistics. , HOME - Project: measure the nexus between environment and gender Project countries: Cambodia Mongolia Resources Regional Events
Organizer(s): SESRIC Economic Cooperation Organization DOSM Malaysia
Description: Within the framework of its Statistical Capacity Building (StatCaB) Programme, SESRIC will organise an Online Training Course on ‘Big Data Applications on Price Intelligence’ in collaboration with the Economic Cooperation Organization (ECO) for the benefit of National Statistical Offices (NSOs) of OIC countries on 14-16 June 2022. Ms. Maslina Samsudin, Deputy Director; Ms. Mazliana Mustapa, Principal Assistant Director; Mr. Wan Ahmad Ridhuan Wan Jaafar, Senior Assistant Director; and Ms. Noradilah Adnan, Assistant Director at the Department of Statistics Malaysia (DOSM), will conduct the course and cover the following topics: Overview on Big Data Compilation of Consumer Price Index (CPI) in Malaysia Price Intelligence Module in Price Intelligence Data Acquisition (Data Crawling) Data Management including Classification and Data Generator Data Visualisation including Demo on Price Intelligence Visualisation using Tableau The course will be conducted through a video conferencing platform by following synchronous learning and instruction approaches designed in line with the virtual training solutions undertaken by SESRIC in order to better serve the Centre’s training activities and keep participants motivated and engaged during this time of global crisis due to COVID-19. For more information on SESRIC Statistical Capacity Building (StatCaB) Programme, please visit: http://www.oicstatcom.org/statcab.php
Organizer(s): UN Global Pulse Ministry for Foreign Affairs Finland
Description: Digital Public Goods are a fundamental pillar in achieving the Sustainable Development Goals and driving the United Nations (UN) Roadmap for Digital Co-operation. UN Global Pulse has been working to identify what innovators need to ensure their products are able to scale successfully. UN Global Pulse and the Ministry for Foreign Affairs of Finland invite you …
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Description: Big data is increasingly making in-roads in the compilation of official statistics. Not only does it ensure timely availability of statistics but is effective in terms of costs of statistics production. ESCAP developed a big data strategy to ably support its member states innovate statistical production systems. Most of the big data can be collected from administrative sources or computer databases where it is continuously generated. Pioneers of big data have often been private sector enterprises that have adopted various technologies such as machine learning to predict statistical outcomes to meet their business needs. While big data has foot in the private sector, official statistical agencies have begun to adopt it in their operations. Statistical offices have realized the need to remain relevant as well as remain leaders in the production of statistics. Similarly, they are beginning to realise that despite lack of structure and standards, big data complements existing statistical products and offers users another set of statistical information to meet their everyday needs. Over time, users of statistics have emphasized the need for timely statistics. Timeliness of statistics remains a critical aspect of data quality. While traditional statistics are reputed for strong adherence to standards, their timeliness has often been in question. Big data lacks structure but its timeliness, gives statistical agencies a critical tradeoff between timeliness and relevance aspects of data quality. It also gives statistical agencies a good foundation to adopt innovation in terms of how data is collected, processed and its ability to address the ever-changing needs of technology savvy users. Price statistics are one area of official statistics that has increasingly adopted the use of big data in its production amongst some member states of ESCAP region1. Price statistics are often collected through price surveys and indices compiled based on household/retail surveys that are often infrequent or outdated. The increased digitalization of retail business offers an alternative to household surveys and if properly done can enhance the accuracy of price statistics. In view of these developments, ESCAP under the umbrella of the steering group of the regional programme for the improvement of economic statistics in Asia and the Pacific, organised an online seminar on the use of big data in price statistics on 21 April 2022. This seminar was organised as part of implementing the ESCAP big data strategy and targets economic statisticians involved in the compilation and use of price statistics in the Asia-Pacific region. The seminar exchanged and discussed examples of big data use for price statistics production. Experts from the World Bank, Statistical Centre of Iran, and Statistics New Zealand made presentations during the seminar. 1https://www.unescap.org/sites/default/d8files/knowledge-products/Stats_Brief_Issue28_Big_data_for_economic_statistics_Mar2021.pdf
Description: During this Global Network Webinar, Prof. Qunli Han, a distinguished professor of the International Research Center of Big Data for Sustainable Development Goals (CBAS), gave a presentation on harnessing big data and geospatial information to support the monitoring of the SDGs. The assistant director of the United Nations Statistics Division and the representative of the Global Geospatial Information Management section provided opening remarks. During the presentation, Prof. Qunli introduced the vision, mission and objectives of CBAS and its ongoing projects toward the monitoring of SDGs, including Goals 2, 6, 11, 13, 14 and 15. It was also highlighted the key digital infrastructure that has been developed for CBAS, including the SDG big data platform and the recently launched SDG satellite (SDGSAT-1). In addition, the webinar discussed the collaboration with international organizations, the involvement of international experts in CBAS, and upcoming projects to monitor other SDGs.
Source: Eurostat (Data extracted on: 30 Jan 2022 )
[+] More
Organizer(s): Eurostat Icon-Institut
Description: Main objectives of the course are: * Introducing the participants to the concept of Big Data, the associated challenges and opportunities, and the statistical methods and IT tools needed to make the use of Big Data effective in official statistics. * Overviewing statistical methods and IT tools available for Big Data usage in Official Statistics.
Target Audience: Official statisticians (including managers) to be involved in big data activities and having no specific knowledge on this subject; Official statisticians (including managers) who, without being directly involved in big data activities, need basic knowledge on the use of big data in official statistics.
Source: Eurostat (Data extracted on: 04 Jan 2022 )
[+] More
Organizer(s): Eurostat Icon-Institut
Description: Objectives: * Validation and reporting results of machine learning methods; * Mathematical concepts of supervised and unsupervised machine and deep learning models, like PCA, SVM, trees, ensembles and neural networks; * Use of scikit-learn, matplotlib, pandas, tensorflow, keras to design models and perform machine learning experiments; * Understanding of symbolic computation for backpropagation and gradient descent; * Model selection, Hyper-parameter tuning and practical considerations; * Understanding the lego bricks of neural networks (deep learning) and the numerical issues, in particular vanishing gradients; * Use pretrained models for text and vision applications with libraries like deeppavlov and detectron2.
Target Audience: Python programmers with ambitions to apply machine and deep learning in software engineering and research questions.
Source: Eurostat (Data extracted on: 04 Jan 2022 )
[+] More
Organizer(s): Eurostat Icon-Institut
Description: Objectives: * This course shall democratize the ethical development and use of AI; * Understand the working principles of Artificial Intelligences (AI); * Get to know the processes and tool involved in creating an AI; * Get to know impactful use cases of AI; * Understand the risks and challenges related to AI; * Create an AI from scratch, without code.
Target Audience: Statistical production units and methodologist of NSIs.
Description: With all these efforts of accelerating the modernization of official statistics, it is still difficult to reach and involve the statistical offices in all corners of the world. To be more inclusive and bring modernization closer to home, four Regional Hubs for Big Data in support of the UN Global Platform were established in Brazil (Rio de Janeiro), China (Hangzhou), Rwanda (Kigali) and UAE (Dubai) during 2020 and 2021. The main objectives of the Regional Hubs are the facilitation of innovation projects, the sharing of methods, algorithms and tools, and the provision of training in the use of Big Data and data science for the community of official statisticians in their respective regions. The Regional Hubs will be used to organize international seminars and workshops on data science and modernization of official statistics. For example, the National Bureau of Statistics of China organized thus far two international seminars in December 2020 and September 2021 in hybrid format, with in-person participation of national statisticians and remote participation of international statisticians.
Organizer(s): BIS ECB Bank of Italy Reserve Bank South Africa
Description: Organised by the IFC at the BIS with the active support of the Bank of Italy, the European Central Bank and the South African Reserve Bank, this workshop will bring together central banks, international organisations, national statistical offices and other interested stakeholders to share knowledge on emerging trends in data science, data engineering and information technologies with a broad audience of practitioners and technicians. We will look in depth at the state of adoption of data analytics and business intelligence techniques along with data transformation and big data ecosystems in participants’ organisations. This event is intended to foster exchange, collaboration and understanding on the related interdisciplinary practices, use cases, and technologies and will also cover important topics such as data governance and data protection. The first part of the workshop will be hosted by the Bank of Italy (BoI) on 19-22 October 2021 and the second part by the Bank for International Settlements (BIS) on 14-17 February 2022. The first part would focus on “Data Science in Central Banking: Machine learning applications” whereas the second part would be on “Data Science in Central Banking: Applications and tools”.
Description: The Federal Competitiveness and Statistics Centre (FCSC) of the UAE together with the United Nations Committee of Experts on Big Data and Data Science for Official Statistics (UN-CEBD) welcome the world to explore data science and big data during a three-day event at Expo 2020. FCSC also invites everyone to witness the ceremonial launch of the Regional Hub for Big Data and Data Science in UAE, which serves the Middle East and North Africa (MENA) Region.
Description: The fourth Road to Expo 2020 will be held on 14 December and will focus on concrete methods for calculation of SDG-indicators by big data. Sustainable Development Goals with their broad coverage of societal, environmental and economic fields pose challenges to a comprehensive statistical follow-up. Possible data sources towards this aim are widely discussed and data from non-traditional sources are frequently mentioned in this regard. In this context, data from non-traditional sources can include big data, geo data or citizen generated data. The webinar will present concrete examples of how selected SDG-indicators can be calculated with the use of non-traditional data, hereby giving countries a concrete tool for reinforcing their statistical follow-up on the SDG. Furthermore, the webinar will discuss future possibilities for monitoring the SDG with non-traditional data, including geo and citizen generated data. Finally, while giving an insight into monitoring the SDG by non-traditional data, the webinar will build up to the workshop during the EXPO 2020, which subsequently will expand the topic and present the calculation methods in detail.
Organizer(s): UN MGCY UNSD UN Global Platform Regional Hubs
Description: Over the last years, the world has been forced to rethink, redesign and innovate new ways of creating a shared sustainable future. This year, together with the United Nations Statistics Division, and the global platform regional hubs of Rwanda, UAE, Brazil & China, MGCY is organizing a virtual UN Youth Hackathon to create solutions to some of the world’s toughest problems using data. During this hackathon, the UNSD, regional hubs and partner organizations will be on the lookout for disruptive and innovative solutions that use data, machine learning and artificial intelligence to fast track progress towards the UN Sustainable Development Goals (SDGs). We are looking for enthusiastic data people - students and professionals in the broad field of AI, Deep Learning and Data Science. There are no strong requirements other than the strong interest in contributing to the UN SDGs.
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Description: For the past two years, UNICEF and the Government of Finland have partnered to better understand how artificial intelligence (AI) systems can protect, provide for and empower children. On 30 November and 1 December 2021, we jointly hosted a virtual Global Forum on AI for Children, which gathered experts, policymakers, practitioners, researchers, children and young people to share their knowledge, expertise and experience on the use of AI systems by and for children. At the event we launched Version 2.0 of the Policy Guidance on AI for Children, and the organizations that have piloted the guidance shared their experiences and lessons learned.
Description: On GEO Week 2021, support the conversation on how Earth Observation can inform policy development and decisions on the ground to ensure no one is left behind.
Source: ESCAP SIAP (Data extracted on: 04 Jan 2022 )
[+] More
Organizer(s): ESCAP ADB ESCAP SIAP
Description: The course is designed for personnel working in the field of statisticswhose main responsibilities include data analysis of SDG indicators and related statistics with a specific target on data scientists from NSOs with an experience in both statistical modelling (regression analysispredictionclassification...) and with programming or algorithmic skills. Although no programming will be required to follow and succeed in the coursethe pedagogical materials include R codein the form of reproducible markdown notebooksas well as some Python resources and code.
Organizer(s): BIS ECB Bank of Italy Reserve Bank South Africa
Description: Organised by the IFC at the BIS with the active support of the Bank of Italy, the European Central Bank and the South African Reserve Bank, this workshop will bring together central banks, international organisations, national statistical offices and other interested stakeholders to share knowledge on emerging trends in data science, data engineering and information technologies with a broad audience of practitioners and technicians. We will look in depth at the state of adoption of data analytics and business intelligence techniques along with data transformation and big data ecosystems in participants’ organisations. This event is intended to foster exchange, collaboration and understanding on the related interdisciplinary practices, use cases, and technologies and will also cover important topics such as data governance and data protection. The first part of the workshop will be hosted by the Bank of Italy (BoI) on 19-22 October 2021 and the second part by the Bank for International Settlements (BIS) on 14-17 February 2022. The first part would focus on “Data Science in Central Banking: Machine learning applications” whereas the second part would be on “Data Science in Central Banking: Applications and tools”.
Description: As part of the EXPO2020, the Federal Competitiveness and Statistics Center of UAE (FCSC) will host a 3-day event in January 2022 organized jointly with the United Nations Committee of Experts on Big Data and Data Science for Official Statistics (UN-CEBD). The event is about data solutions that will support the implementation of the Sustainable Development Goals and help in overcoming emergency situations such as the COVID-19 pandemic, for example, by using mobile phone data. It will showcase collaboration opportunities on the UN Global Platform and the Regional Hubs for Big Data, especially the Regional Hub in Dubai. From September to December 2021, UN-CEBD will organize a series of webinars in close collaboration with FCSC/UAE, paving the road to the event. The first of these webinars will be held on 28 September and will focus on the UN Global Platform and show several projects currently running on the platform.
Description: Eurostat together with the Scanner Data Task Team of the UN Committee of Experts on Big Data and Data Science for Official Statistics will organize a workshop on scanner data and web scraping on 12-14 October 2021. The objectives of the workshop are to exchange on practical experiences related to the use of new data sources in consumer price statistics, and to present and discuss some of the work conducted by the Task Team and by the European Statistical System.
Source: Eurostat (Data extracted on: 21 Dec 2020 )
[+] More
Organizer(s): Eurostat Icon-Institut
Description: Learn how to extract relevant information for statistical purposes from huge amounts of data.
Target Audience: IT professionals whose role is to support statisticians with big data infrastructure, either via local big data clusters or via cloud solutions, and the engineering of big data processing. Methodologists and statisticians with a strong IT background who are expected to handle big data infrastructures and unstructured data. ESTP Trainings are open to non-ESS members if capacity allows after ESS needs are fulfilled.
Organizer(s): Qatar Computing Research Institute MIT United States Max Planck Institute for Demographic Research Germany
Description: Billions of people use social media platforms. An important aspect that social media provides is the real time access to digital trace data. This offers complementary strengths to existing, expensive and typically slow paced data from surveys (e.g. migration statistics or disaster impact assessment). In particular the need for near real-time data analysis that does not rely on face-to-face interactions has been made clear by the current pandemic. UN Global Pulse will speak at the event about the innovative use…
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Source: Eurostat (Data extracted on: 03 May 2021 )
[+] More
Organizer(s): Eurostat Icon-Institut
Description: Providing knowledge and basic practice on use of earth observation data to produce statistics, EO data sets, processing analyses and presenting.
Target Audience: NSIs and Other National Authorities (with lower priority), working in the field of Earth observation, geographic information systems and statistics. ESTP Trainings are open to non-ESS members if capacity allows after ESS needs are fulfilled.
Description: The seminar is jointly organized by the National Bureau of Statistics of China (NBS), the United Nations Statistics Division (UNSD), and the United Nations Global Platform for Big Data China Hub.
Description: This Stats Café explored different big data partnership models with government agencies and the private sector, touching upon data privacy issues, necessary legal adjustments to accessing and using big data for official statistics, and win-win models for data exchange. Related events Asia-Pacific Stats Café Series: Big Data Governance, 16 August 2021. Expert Group Meeting on Big Data for Official Statistics: Big Data Governance (Expert Discussion I), 19 August 2021, 2 September 2021 (Read Report) , Stats Café Home: Upcoming events Concluded events
Title in Arabic: رسم خرائط الطائرات بدون طيار ونظام المعلومات الجغرافية والذكاء الاصطناعي
Organizer(s): AITRS
Description: أصبحت خرائط الطائرات بدون طيار "أداة" قوية لإنشاء خرائط ثنائية الأبعاد ونماذج ثلاثية الأبعاد يمكن استخدامها في أنواع مختلفة من التطبيقات من الزراعة ورسم الخرائط والطاقة والبناء وحتى الاستجابة للطوارئ. ويتيح دمج خرائط الطائرات بدون طيار التي تعمل بالذكاء الاصطناعي مع نظام المعلومات الجغرافية (GIS) وأجهزة الاستشعار الحصول على بيانات ومعلومات تمكن من إجراء تحليلات وتنبؤات أكثر دقة واتخاذ قرارات أفضل. وتهدف هذه الورشة التدريبية الى تمكين المشاركين من التعرف على متطلبات تقنيات رسم خرائط بالطائرات بدون طيار وكيفية انجاز العمل الميداني بهذه التقنية، بالإضافة الى كيفية دمج المعلومات المتحصل عليها مع ما تتيحة وأنظمة المعلومات الجغرافية والاستشعار عن بعد وتوفير بيانات احصائية تساعد على التحليل واتخاذ القرارات المبنية على الادلة وخاصة في إطار اهداف التنمية المستدامة. وفي ما يلي تقديم أكثر للورشة باللغة الإنجليزية نظرا لصعوبة ترجمة بعض المصطلحات ولمزيد فهم المطلوب في هذا المجال الجديد
Description: The Asia-Pacific Stats Café Series on "Big Data Governance" was held on Monday, 16 August 2021, 12:00-13:30 hours (GMT+7). Attendance summary Flyer This Stats Café explored governance issues related to the use of big data for official statistics, including privacy, ethics, legislation and coordination, and how the role of the NSOs in government-wide digital strategies, national data infrastructure and data sharing can be leveraged for expanded uptake of big data in statistical operations. Related events Asia-Pacific Stats Café Series: Big Data partnership models, 30 August 2021. Expert Group Meeting on Big Data for Official Statistics: Big Data Governance (Expert Discussion I), 19 August 2021, 2 September 2021 (Read Report) , Stats Café Home: Upcoming events Concluded events
Description: The Asia-Pacific Stats Café Series on "Machine Learning for Sentiment Analysis" was held on Monday, 26 July 2021, 12:00-13:00 hours, Bangkok time (GMT+7). Attendance summary Flyer , Stats Café Home: Upcoming events Concluded events
Description: The Federal Statistical Office (FSO) of Switzerland together with the United Nations Statistics Division (UNSD) and Facebook are organizing a webinar on the topic of Data Science and Official Statistics. FSO recently established a Data Science Competence Centre which will benefit all government departments of Switzerland, while UNSD has been supporting the UN Committee of Experts on Big Data and Data Science for Official Statistics since 2014. Facebook uses Data Science in developing market strategy, but also to support a large Open Source community, called PyTorch, which has developed a machine learning framework that accelerates the path from research prototyping to production deployment.
Description: Success in data science and statistics is dependent on the development of both analytical and computational skills, and the demand for educators who are proficient at teaching both these skills is growing. The goal of this workshop is to equip educators with concrete information on content, workflows, and infrastructure for painlessly introducing modern computation with R and RStudio within a data science curriculum. In addition to gaining technical knowledge, participants will engage in discussion around the decisions that go into developing a data science curriculum and choosing workflows and infrastructure that best support the curriculum and allow for scalability. Workshop attendees will work through several exercises from existing courses and get first-hand experience with using relevant tool-chains and techniques, including running a course on RStudio Cloud, and literate programming with R Markdown, and workflows for collaboration, version control, and automated feedback with Git/GitHub. We will also discuss best practices for configuring and deploying classroom infrastructures to support these tools.
Target Audience: This workshop is aimed primarily at participants teaching data science in an academic setting in semester-long courses, however much of the information and tooling we introduce is applicable for shorter teaching experiences like workshops and bootcamps as well. Basic knowledge of R is assumed and familiarity with Git is preferred.
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Description: This 2-day virtual course is based on the Data Science and Predictive Analytics (DSPA) course that the instructor teaches at the University of Michigan. The training will provide intermediate to advanced learners with a solid data science foundation to address challenges related to collecting, managing, processing, interrogating, analyzing and interpreting complex health and biomedical datasets using R. Participants will gain skills and acquire a tool-chest of methods, software tools, and protocols that can be applied to a broad spectrum of Big Data problems. Before diving into the mathematical algorithms, statistical computing methods, software tools, and health analytics, we will discuss a number of driving motivational problems. These will ground all the subsequent scientific discussions, data modeling, and computational approaches. The training will involve active-learning and integrate driving motivational challenges with mathematical foundations, computational statistics, and modern scientific inference. Building on open-science principles, training will focus on effective, reliable, reproducible, and transformative data-driven discovery. Trainees will develop scientific intuition, computational skills, and data-wrangling abilities to tackle Big biomedical and advanced health data problems. The instructor will provide well-documented R-scripts and software recipes implementing atomic data-filters, as well as complex end-to-end predictive big data analytics solutions.
Target Audience: Intermediate to advanced level learners, e.g., graduate students, postdocs, fellows, data science practitioners, engineers, mathematical modelers, (technology) team leaders, health analysts. The prerequisites incluyde some college-level quantitative training.
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Description: This course describes nonparametric regression, including the additive model and its generalizations, also the LASSO, and LARS. Then it proceeds to classification (SVMs, random forests, boosting). The Curse of Dimensionality is described in both contexts. Bagging and stacking are covered. There is some coverage of cluster analysis, and some text analytics. The emphasis is upon the strengths and weaknesses of the tools, and guidance on when a particular method should be used.
Target Audience: People who are familiar with multiple linear regression. I expect someone who got a MS in statistics ten years ago would be a good fit. People with stronger backgrounds will still benefit, and probably acquire deeper insights.
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Description: During this Global Network Webinar, we were happy to have Andrew Trask from OpenMinded, who provided A Tutorial on Remote Data Science. The presentation walked through the main tools (federated learning, differential privacy, etc.) for remote data science, their current state of maturity, and opinions on where these tools will take our society in the next five years. These tools offer the promise of instant, safe, privacy-preserving access to data across every organization with data relevant to your problem.
Organizer(s): SESRIC ESCWA Statistics Indonesia DOS Jordan DOSM Malaysia NCSI Oman Statistics Saudi Arabia TurkStat Türkiye
Description: Within the framework of its Webinar Series on Statistical Experience Sharing, SESRIC will organise a webinar on ‘Big Data Applications and Utilising Non-Traditional Data Sources and Methods for Official Statistics’ in collaboration with National Statistical Offices (NSOs) of Indonesia, Jordan, Malaysia, Oman, Saudi Arabia, Turkey, and the Statistics Division of the United Nations Economic and Social Commission for Western Asia (UNESCWA) on 10 June 2021 with the participation of NSOs of OIC countries. The objective of the webinar is to share experiences in the development and/or use of big data applications and utilising non-traditional data sources and methods for official statistics. The webinar will cover the following topics: Opportunities and challenges of using non-traditional data sources, particularly big data, for official statistics; and Concrete big data applications of the OIC countries including the use of non-traditional data sources and methods such as satellite imagery data, mobile phone data, scanner data, Automated Identification System (AIS) vessel tracking data, earth observation data, machine learning, and others. The webinar will be conducted through a video conferencing platform by following synchronous learning and instruction approaches designed in line with the virtual training solutions undertaken by SESRIC in order to better serve the Centre’s training activities and keep participants motivated and engaged during this time of global crisis due to COVID-19. Documents: Concept Note (English)
Description: This course covers foundation and recent advances of deep neural networks (DNNs) from the point of view of statistical theory. Understanding the power of DNNs theoretically is arguably one of the greatest problems in machine learning. During the last decades DNNs have made rapid process in various machine learning tasks like image and speech recognition and game intelligence. Unfortunately, little is yet known about why this method is so successful in practical applications. Recently, there are different research topics to also prove the power of DNNs from a theoretical point of view. From an aspect of statistical theory, several results could already show good convergence results for DNNs in learning different function classes. The course is roughly divided into two parts. In the first part, DNNs are introduced and different network architectures are discussed. In the second part, we focus on the statistical theory of DNNs. Here we will introduce frameworks addressing two key puzzles of DNNs: approximation theory, where we gain insights in the approximation properties of DNNs in terms of network depth and width for various function classes and generalization, where we analyze the rate of convergence for both, regression and classification problems.
Target Audience: PhD students and researchers in the area of mathematics/statistics/computer science.
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Description: Spatial data science is concerned with analyzing the spatial distributions, patterns, and relationships of data over a predefined geographical region. It relies on the dependence of observations where the primary assumption is that nearby spatial values are associated in a certain way. For decades, the size of most spatial datasets was modest enough to be handled by exact inference. Nowadays, with the explosive increase of data volumes, High-Performance Computing (HPC) has become a popular tool for many spatial applications to handle massive datasets. Big data processing becomes feasible with the availability of parallel processing hardware systems such as shared and distributed memory, multiprocessors and GPU accelerators. In spatial statistics, parallel and distributed computing can alleviate the computational and memory restrictions in large-scale Gaussian random process inference. In this course, we will first briefly cover the motivation, history, and recent developments of statistical methods so that the students can have a general overview of spatial statistics. Then, the cutting-edge HPC techniques and their application in solving large-scale spatial problems with the new software ExaGeoStat will be presented.
Target Audience: Statisticians with interests in High-Performance Computing and large-scale Spatial Statistics.
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Organizer(s): University of Manchester United Kingdon
Description: Big data, which is large datasets generated from digital devices and sensors (e.g. social media, mobile data and remote sensing data) are creating new possibilities to transform development research and policy. However, such data are usually decontextualized data which decrease the meaning and value that can be extracted from it. More traditional thick data collected through qualitative and ethnographic methods could rescue big data from this context-loss as they are better able to explain the why and how of what…
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Description: The NCSI Oman has accomplished a major achievement by using mobile positioning data for official statistics and making it one of the important data sources for the measurement of tourism, population, and commuting in the Sultanate of Oman. The use of mobile positioning data allows to reduce the volume of field surveys, reduce frequent visits to families and companies to conduct field surveys, and increase the speed of response to the requirements of development plans. In addition, it can be used to cover new statistical areas that development plans may require. This achievement of using Big Data is a paradigm shift in the production processes of high-quality official indicators.
Description: This is a side-event of Twentieth meeting of the Executive Committee of the Statistical Conference of the Americas of ECLAC, organized by INEGI of Mexico.
Title in Arabic: اسس ومفاهيم البيانات الضخمة والذكاء الاصطناعي
Organizer(s): AITRS UNDP Arab Development Portal Arab Aid
Description: ازدادت أهمية علم البيانات مؤخراً بشكل كبير نظراً لحجم البيانات الضخمة المتوفرة من مصادر متعددة وبأشكال مختلفة وهو ما يعرف بمصطلح البيانات الضخمة Big Data وحاجة هذه البيانات إلى طرق وأدوات معالجة خاصة. وقد طال الاهتمام بهذا العلم أجهزة الاحصاء الرسمية تجلى أساسا في عقد مؤتمرات وورش عمل اقليمية ودولية تتعلق بالبيانات الضخمة وعلم البيانات، والاستغلال الأمثل لما تتيحه هذه التكنولوجيات في تطوير الاحصائيات الرسمية. كما شرع العديد من الاجهزة الاحصائية في التطبيق الفعلي لمختلف التطورات والتطبيقات الجديدة في هذا المجال. وفي هذا السياق، سيتم تنظيم مجموعة ورشات عمل إقليمية مترابطة ومتكاملة عن بعد للأجهزة الإحصائية في البلدان العربية حول علم البيانات والذكاء الاصطناعي للتعريف بهذا الموضوع الهام . أهداف الورشات وتتمثل أهداف الورشات أساسا في: * إطلاع المشاركين على مفهوم علم البيانات، والبيانات الضخمة باستخدام تقنيات الذكاء الاصطناعي * إدراك المشاركين لأهمية علم البيانات في الاحصاءات الرسمية والتعرف على مختلف مكوناته، * التعرف على مختلف الاستعمالات والقيمة المضافة للبيانات الضخمة والذكاء الاصطناعي، * إطلاع المشاركين على الطرق الأولية في معالجة البيانات باستخدام برامج محددة.
Description: This side-event will showcase the competence framework on Big Data and data science and the related maturity matrix for statistical offices, the training program on Earth observations for agriculture statistics and the training courses on privacy preserving techniques. The objective is to convey the message that training in Big Data and data science is being developed and is available and open to all the members of the Commission.
Description: In the context of the Global Working Group on Big Data and its Task Team on Big Data for the SDGs in particular, the purpose of this side-event is to share status of the current work on monitoring the SDGs with the help of non-traditional data, including big data sources such as Earth observations and citizen generated data. The aim of the side event is to provide inspiration and knowledge exchange for countries wishing to exploit big data for their SDG reporting.
Source: Eurostat (Data extracted on: 21 Dec 2020 )
[+] More
Organizer(s): Eurostat Icon-Institut
Description: Objectives: * Validation and reporting results of machine learning methods. * Mathematical concepts of supervised and unsupervised machine and deep learning models, like PCA, SVM, trees, ensembles and neural networks. * Use of scikit-learn, matplotlib, pandas, tensorflow, keras to design models and perform machine learning experiments. * Understanding of symbolic computation for backpropagation and gradient descent. * Model selection, Hyper-parameter tuning and practical considerations. * Understanding the lego bricks of neural networks (deep learning) and the numerical issues, in particular vanishing gradients. * Use pretrained models for text and vision applications with libraries like deeppavlov and detectron2.
Target Audience: Python programmers with ambitions to apply machine and deep learning in software engineering and research questions. ESTP Trainings are open to non-ESS members if capacity allows after ESS needs are fulfilled.
Description: About the session The UN Statistical Commission’s Working Group on Open Data provides good practice guidance on geographical areas, content and the balance between openness and privacy for local-level statistics. For many countries, this balance is often tipped towards closure not openness and therefore limiting their ability to engage the public and public policy makers with data and statistics that matter to them. This side event shared use cases from the Netherlands, New Zealand and Mexico. Three countries where a balance has been successfully struck. The event also provided good practice guidance on frameworks for striking a balance such as Statistics Canada’s data model vision for a National Statistical System (GATHER-GUARD-GROW-GIVE).
Description: This side-event of the 52nd UN Statistical Commission will showcase the projects on Sen2Agri, AIS data and .STAT. The objective is to convey the message that the data, services, technology and applications on the platform are available and open to all the members of the Commission. More projects can be onboarded. The side event will be conducted virtually and showcase 3 presentations with Q&A.
Source: Eurostat (Data extracted on: 21 Dec 2020 )
[+] More
Organizer(s): Eurostat Icon-Institut
Description: Main objectives of the course are: * Introducing the participants to the concept of Big Data, the associated challenges and opportunities, and the statistical methods and IT tools needed to make the use of Big Data effective in official statistics. * Overviewing statistical methods and IT tools available for Big Data usage in Official Statistics.
Target Audience: Official statisticians (including managers) to be involved in big data activities and having no specific knowledge on this subject; Official statisticians (including managers) who, without being directly involved in big data activities, need basic knowledge on the use of big data in official statistics. ESTP Trainings are open to non-ESS members if capacity allows after ESS needs are fulfilled.
Description: About the session This session explored how the NSOs use big data and data science in the production and improvement of economic indicators (i.e., scanner data and web scraping for the CPI) and how they develop data partnerships with the private sector. Agenda Welcome remarks: Tanja Sejersen, Statistician, Statistics Division, UN ESCAP Speakers: Mazliana Mustapa, Principal Assistant Director, Department of Statistics, Malaysia Gary Dunnet, Deputy Chief Methodologist, Statistics New Zealand Tom Smith, Director, Data Science Campus, UK Office for National Statistics Moderator: Irina Bernal, Consultant, UN ESCAP --------------------- >> See others Asia-Pacific Stats Café series
Description: Agenda Opening remarks: Ms. Gemma Van Halderen, Director, Statistics Division, UN ESCAP Introduction: Arturo Martinez, Statistician, Statistics and Data Innovation Unit, ADB GeoSpatial Data for Agriculture at the Ministry of Agriculture and Rural Development of Viet Nam: Dr. Do Minh Phuong, Senior Researcher, Ministry of Agriculture and Rural Development, Viet Nam Mapping poverty through Data Integration and Artificial Intelligence: Ms. Anna Jean C. Pascasio, Senior Statistical Specialist at the Poverty and Human Development Statistics Division of the Philippine Statistics Authority and Ms. Budsara Sangaroon, Group Director of Statistical Analysis in Social Sector of the Statistical Forecasting Division of the National Statistical Office of Thailand Discussant: Dr. Romulo Virola, Secretary General of the former National Statistical Coordination Board of the Philippines Q/A and discussion: moderated by Mr. Arturo Martinez. Relevant Publications: Earth Observations for a Transforming Asia and Pacific Use of Remote Sensing to Estimate Paddy Area and Production – A Handbook Mapping Poverty Through Data Integration and Artificial Intelligence
Big data tools for IT professionals supporting statisticians
09 Nov 2020 – 30 Nov 2020
Source: Eurostat (Data extracted on: 17 Nov 2020 )
[+] More
Organizer(s): Eurostat Icon-Institut
Description: Learn how to extract relevant information for statistical purposes from huge amounts of data.
Target Audience: IT professionals whose role is to support statisticians with big data infrastructure, either via local big data clusters or via cloud solutions, and the engineering of big data processing. Methodologists and statisticians with a strong IT background and who are expected to handle big data infrastructure on their own. ESTP Trainings are open to non-ESS members if capacity allows after ESS needs are fulfilled.
Inaugural BD4M Webinar Series: Estimating migrant stocks and flows using social media data
03 Nov 2020 – 03 Nov 2020
Source: IOM GMDAC (Data extracted on: 18 Nov 2020 )
[+] More
Organizer(s): IOM GMDAC EU New York University United States
Description: "Estimating migrant stocks and flows using social media data" is the inaugural webinar of the BD4M Webinar Series: Harnessing Data Innovation for Migration Policy.
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Description: About the session The use of modern technologies, such as mobile phones or social media generate data about people’s movements, activities and opinions. When analyzed in a privacy-preserving manner, these data could translate into valuable insights and statistics to inform public policy. National statistical offices in Asia-Pacific region have been exploring ways of generating social and demographic statistics from big data sources, with most pilots focusing on human mobility and migration statistics. This session showcased several country examples in Asia-Pacific region of compiling social and demographic statistics using big data, in particular the mobile phone data. The discussions focused on data access and privacy, big data integration, and scaling up big data pilots. Speakers Edi Setiawan, Head of Population & Labor Mobility, Statistics Division, BPS-Statistics Indonesia Rajius Idzalika, Data Scientist, Pulse Lab Jakarta Linus Bengtsson, Executive Director, Flowminder Foundation Tanja B. Sejersen, Statistician, Statistics Division, ESCAP --------------------- >> See others Asia-Pacific Stats Café series
Description: Data algorithms are critical tools that have brought benefits to the public and private sector and the broader Data Community. However, public trust in artificial intelligence-based Data Algorithms has faltered due to a lack of transparency in how Data Algorithms function and public involvement in decision making around how such Algorithms are used. In line with recommendation 3C of the report of the UN Secretary-General’s High-Level Panel on Digital Cooperation, the session will discuss challenges and solutions relating to the…
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Description: The UN Global Working Group (GWG) on Big Data for Official Statistics will conduct a webinar jointly with the Federal Statistical Office (FSO) of Switzerland on the topic of data innovation in official statistics. FSO has taken the initiative to include data science into the portfolio of its work. This will be for the benefit not only of the Federal Statistical Office but also of other government departments. The GWG just conducted a Hackathon in the use of AIS data on the UN Global Platform. AIS is real-time data which tracks the position of ships around the world, and the UN Global Platform is a cloud-based collaborative environment for the GWG community of statisticians and data scientists. During the Hackathon 17 teams used AIS data to develop new applications, which can help in the COVID-19 response or in Climate Action. The winner and runner-up of this contest will present their application during this webinar.
Description: About the session Earth Observation (EO) data constitute an invaluable resource for complementing traditional data sources for official statistics in support of the 2030 Agenda and more countries are integrating EO data into the production of official statistics. This session highlighted country examples from Asia-Pacific in producing environment-related official statistics using Earth Observation data. Presentations focused on methodologies, SEEA and SDG indicators, data sharing and collaboration across institutions, and the opportunities presented by Open Data Cubes for official statistics. In addition, ESCAP launched the Guideline on Land Cover Statistics using satellite data. Agenda: Welcome: Rikke Munk Hansen, Chief of the Economic and Environment Statistics Section, UNESCAP Launch of the Guide on Producing land cover change maps and statistics Compiling Ocean Account using Earth Observation data in Viet Nam: Viet Anh Hoang Open Data Cube at the service of the statistics community in the Pacific: Stuart Minchin Use of satellite data for official statistics: Gemma van Halderen Questions and answers: Rikke Munk Hansen Speakers: Viet Anh Hoang is a Researcher at the Research Institute of Forest Ecology and Environment, Viet Nam Academy of Forest Science with over 20 years of experience in forestry and GIS research Stuart Minchin is the Director-General of the Pacific Community (SPC) and previously served as Chief of the Environmental Geoscience Division of Geoscience Australia Gemma van Halderen in the Director of Statistics Division, ESCAP --------------------- >> See others Asia-Pacific Stats Café series
Organizer(s): UNSD UNCTAD UN Global Pulse MarineTraffic CCRi
Description: Automatic identification system (AIS) data allows for real-time geo-tracking and identification for equipped vessels. AIS data provides a big data source of unrivaled quality. The number of possible applications for this data is enormous. To quickly utilize the data to its full potential, we need a surge of creative researchers to come up with equally creative ways to use it!
Description: Governments and private actors are leveraging the power of big data and Artificial Intelligence (AI) to monitor and contain the spread of the coronavirus (COVID-19). Secure, timely and reliable data collection and sharing on a global scale are critical to understanding how the virus spreads, maximizing the effectiveness of government policies, and promoting international cooperation in the race to create and distribute therapies and vaccines. A demand for closer, forward-looking international collaboration has emerged from the current crisis. Yet, national…
Topics:
Original webpage was deleted, archived version from the Internet Archive (not a UN service): Link
Description: The focus of this year's conference is on the use of Data Science for official statistics, in particular the use of artificial intelligence (AI) and machine learning. In addition to the creation of Big Data centers next to the national statistical offices (NSOs) they have started creating Data Science Campuses as R&D centers, which develop and test new methods. They also have an educational program usually in cooperation with some universities. The Conference will cover the use of AI and machine learning for official statistics. The UN Global Working Group (UN GWG) created the UN Global Platform as a collaborative environment to develop and test new data sources, methods and algorithms for the global statistical system, such as new methods to compile agricultural crop yield using satellite data, migration and tourism statistics using mobile phone data and CPI estimates using scanner data which will be discussed during the conference. Many more pilot projects can be initiated on the Platform involving not only the statistical community and related public sector but also stakeholders from private sector, academia and civil society. The Conference will also cover the topic of changing the production line to introduce new data solutions in the national statistical systems.
Description: About the session Big data and other non-traditional data sources are an exciting prospect for national statistical offices faced with increasing demand for more statistics of greater timeliness and relevance as has been experienced in the context of the 2030 Agenda for Sustainable Development and the COVID-19 pandemic. This increasing demand often coincides with declining budgets. Big data and other non-traditional data sources can sometimes offer comparatively inexpensive solutions. Globally and regionally, activities associated with these data sources continue to grow. The opportunities for member State engagement and involvement are vast. This seminar is held as preparation for the 7th Committee on Statistics. The presentations will highlight some of the key global and regional developments to support NSOs to use big data and participants will have a chance to discuss the prospects of big data for official statistics in Asia and the Pacific. Agenda Welcome: Gemma Van Halderen, Director, Statistics Division, ESCAP Big Data and official statistics – the role of the Global Working Group: Niels Ploug Introduction to UN global platform regional hub in China": Wang Wenna Current progress of Big Data utilization for official statistics in Indonesia: Ali Said Questions and answers : Tanja Sejersen Speakers Niels Ploug is the Director of Social Statistics, Statistics Denmark and responsible for SDG data. He is also a cochair UN Global Working Group on Big Data for Official Statistics Wang Wenna is the Head of the Centre on the Application of Big Data in Hangzhou, and Deputy Director-General of NBS Survey Office in Zhejiang, National Bureau of Statistics of China Ali Said is the Director of Analysis and Development of Statistics at Statistics Indonesia. He is also appointed to lead the development of Big Data uses for official statistics at Statistics Indonesia Tanja Sejersen, Statistician, Statistics Division ESCAP --------------------- >> See others Asia-Pacific Stats Café series
Description: This Stats Café featured: Analysis of land cover change for sustainable development Land cover change automation Q&A and panel discussion Moderator Rikke Munk Hansen is the Chief of Economic and Environment Statistics at ESCAP and has worked with statistics and sustainable development in Asia and the Pacific for more than 20 years. Presenters Chitrini Mozumder is Affiliated Faculty and Research Scientist at Asian Institute of Technology with research interests including remote sensing and GIS applications in environmental monitoring and management, land use change analysis and modelling, spatial analysis and decision support. Aahlaad Musunuru is working as a Geographic Information Systems (GIS) specialist in ESCAP-Statistics with skills and research interest in remote sensing and GIS applications for monitoring land cover change, ocean accounts and disaster risk assessment. --------------------- >> See others Asia-Pacific Stats Café series
Description: Digital techniques, such as machine learning (ML), have the potential to add value to statistical data production by not only increasing efficiency in statistical business processes but also allowing the use of new data sources In the production of official statistics, ML techniques could be used for imputation of missing data, reweighting, prediction, and/or calibration to standard classifications. For example, sources of sampling frames, such as administrative data, censuses and other surveys, may be combined through record linkage processes by capitalizing on clustering algorithms to improve the quality of design information on the frame. ML methods, such as regression algorithms, can also be used to predict the probability of response for individual units using information for the entire sample to manage data collection efforts efficiently. These techniques offer many opportunities to improve the efficiency in producing official statistics. National Statistics Offices are also expanding to use of non-traditional sources, such as social media data and mobility data, however, the use of these big data sources come with challenges as they may not meet the requirements of traditional quality frameworks. ML techniques, such as Deep Learning, can be used to tackle this problem, but should be used with caution since these methods can introduce biases due to their low sensitivity to outliers and erroneous data compared to that of classical statistical methods. This webinar will: Showcase examples of ML methods to improve the efficiency of statistical data production; Highlight the use of ML techniques to expand the use of big data sources in official statistics and; Present challenges in using these techniques. This webinar is part of the UN World Data Forum series and highlights the use of non-traditional methods and techniques to add value to official statistics. Speakers Moderator: Steven Vale, Regional Adviser in Statistics, United Nations Economic Commission for Europe (UNECE) Panelist: Jenny Pocknee, Principle Data Scientist, Methodology Transformation Branch, Methodology Division, Australian Bureau of Statistics Presentation Panelist: Alejandro Ruiz, Researcher at National Institute of Statistics and Geography (INEGI) Presentation Speakers' bios
Description: UNWTO, IFC and Digital Tourism Think Tank are providing a 2hr primer masterclass. The executive-level class will provide case studies, quick wins and good practices of using big data in tourism recovery strategies. In this Introductory class, executives will learn: * What big data insights can show us about COVID-19 and the tourism market in Africa; * The spectrum of data opportunities in Africa, who the main providers are and what they offer; * The pros and cons of different uses and approaches to using traditional and big data; * Lessons learned from Portugal, Ethiopia and South Africa.
Description: يتعلق علم دراسة البيانات باستخراج وتحليل وتصور وإدارة وتخزين البيانات لإنشاء رؤى واستخراج معلومات وأفكار منها. ويساعد الدول والمؤسسات على اتخاذ قرارات قوية تعتمد على الحقائق. ويتم استخدام كل من البيانات المهيكلة وغير المهيكلة. إنه مجال متعدد التخصصات له جذوره في الإحصاء والرياضيات وعلوم الكمبيوتر. وتعتبر وظيفة عالم البيانات واحدة من أكثر الوظائف التي يتم البحث عنها بسبب وفرة الطلب وارتفاع الأجور، كما أنها الوظيفة الأسرع نموًا في Linkedin، ومن المتوقع أن توفر 11.5 مليون وظيفة بحلول عام 2026. وهذا يجعل من علم البيانات قطاع عالي التوظيف حيث تعتبر وظيفة عالم البيانات واحدة من أكثر الوظائف ذات الأجور المرتفعة.
