About this short course

This course equips statisticians, researchers, and data practitioners with practical and responsible ways to apply generative artificial intelligence (AI) tools like ChatGPT and Gemini in statistical research and data science. Participants will learn to integrate generative AI across all stages of research, including brainstorming, instrument development, data analysis, and visualization, through hands-on exercises with open datasets like census data and public surveys.

Key topics include foundational concepts of generative AI, such as large language models (LLMs) and generative adversarial networks (GANs), and their strengths and limitations. The course emphasizes practical applications like exploratory data analysis (EDA), statistical inference, visualization, multivariate analysis, and synthetic data generation. Participants will gain hands-on experience using generative AI to uncover patterns, simulate scenarios, and create compelling visuals.

The course also addresses ethical considerations, including bias, hallucinations, and responsible use of generative outputs, ensuring participants apply these technologies thoughtfully. Through interactive demonstrations, practical exercises, and guided discussions, participants will develop the skills to integrate generative AI into their workflows, confidently leveraging its potential to tackle complex data challenges and derive meaningful insights. Evaluation includes pre- and post-course surveys, quizzes, and feedback forms to measure learning outcomes and refine content.

Instructors' biographies:

Dragan Vukmirović, PhD

University of Belgrade, Faculty of Organisational Science, Belgrade, Serbia. Full Professor.

Team leader and key expert in official statistics, former Director General of the Statistical Office of Serbia. Authored over 100 publications in applied statistical research, with recent work focusing on generative AI. Conducted numerous workshops on business analytics, data science, and the implementation of generative AI in business. Holds accreditation for a "Generative AI in Education" course in Serbia.

Tijana Čomić, PhD

University Business Academy in Novi Sad, Faculty of Organisational Studies "Eduka" Belgrade, Serbia, associate professor.

An experienced consultant with UNICEF, specializing in household research and recognized as an international expert on Multiple Indicator Cluster Surveys (MICS). Expertise encompasses Gender, Development, Smart Indicators (GEDSI), and the European Union Statistics on Income and Living Conditions (EU-SILC). Over the past two years, Tijana Čomić has been actively implementing generative AI in their professional practice and leading educational workshops in this emerging field.

For whom is this course intended

This course is designed for statisticians, data scientists, researchers, and data practitioners who wish to explore generative AI's practical and responsible applications in statistics and data science. The target audience includes professionals and academics from diverse fields and graduate students eager to enhance their analytical skills using cutting-edge AI technologies.

Prerequisites:

• A basic understanding of statistical concepts (e.g., probability, hypothesis testing, and data distributions).
• Familiarity with programming in R or Python (essential script execution and familiarity with packages).
• No prior experience with machine learning or AI is required, as the course is structured to provide an accessible introduction to generative AI technologies.

This course is suitable for participants at varying levels of expertise. It focuses on practical applications and hands-on exercises to ensure accessibility and relevance to all.