Numérique - Systèmes d'Information

27210053 - Understanding users for optimal decision-making

Niveau de diplôme
Crédits ECTS 6
Volume horaire total 40
Volume horaire CM 40

Responsables

  • BONHOURE Timothé

Objectifs

This module provides the fundamentals of statistical data analysis, drawing on theories from cognitive psychology and human factors, to support the identification and interpretation of user behavior from raw data. Students will learn to complement large-scale behavioral log data—typically used in data science—with qualitative insights. They will develop a solid grounding in statistics and data-driven evaluation techniques, including A/B testing and selected machine learning approaches.

Estimation of private study (outside of contact hours): 20 hours

TARGETED KNOWLEDGE AND SKILLS 

Knowledge
At the end of this course, students will have acquired:
· A solid understanding of human factors theories (motivation, cognitive load, UX models) relevant to system design and evaluation
· Key concepts in statistical inference, including p-values, effect sizes, and multiple testing corrections (e.g., Benjamini-Hochberg)
· The foundations of behavioral data analysis, combining qualitative and quantitative approaches
· Knowledge of model comparison criteria (BIC, AIC) and data-driven evaluation protocols (e.g., A/B testing)
· Familiarity with standard user self-report instruments (e.g., NASA-TLX, UES, EME)

Skills
By the end of the course, students will be able to:
· Collect, preprocess, and analyze user traces using appropriate statistical and machine learning tools
· Apply machine learning pipelines (training/validation/test splits) for user behavior modeling
· Critically evaluate user experiences and system performance through quantitative and qualitative lenses
· Interpret and report findings in a way that informs decision-making for intelligent system design or strategic improvement

Contenu

COURSE OUTLINE

Session 1 – Psychological Foundations & Motivation Theories
Session 2 – User behavior analysis? From log data to insight
Session 3 – Designing & Critiquing Self-Report Measures
Session 4 – Trace-Based Modeling of User Behavior
Session 5 – Experimental Design & A/B Testing
Session 6 – Statistical Testing for Behavioral Data
Session 7 – Modeling User Behavior with ML
Session 8 – Evaluating and Comparing Models
Session 9 – Aggregation & Triangulation of User Data
Session 10 – Project Presentations & Discussion

Bibliographie

PRESCRIBED TEXTS AND PUBLICATIONS

1. Ryan, R. M., & Deci, E. L. (2008). Self-determination theory and the role of basic psychological needs in personality and the organization of behavior. In O. P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory and research (3rd ed., pp. 654–678). The Guilford Press.
2. Judd, Charles & Mcclelland, Gary & Ryan, Carey. (2017). Data Analysis: A Model Comparison Approach to Regression, ANOVA, and Beyond. 10.4324/9781315744131. 

RECOMMENDED TEXTS AND PUBLICATIONS

3. Peters, D., Calvo, R. A., and Ryan, R. M. (2018). Designing for Motivation, Engagement and Wellbeing in Digital Experience. Frontiers in Psychology, 9.
4. Bouvier, P., Sehaba, K., and LavouÅLe, (2014b). A Trace-Based Approach to Identifying Users’ Engagement and Qualifying Their Engaged-Behaviours in Interactive Systems: Application to a Social Game. User Modeling and User-Adapted Interaction, 24(5):413

Contrôles des connaissances

Individual grade
In class exam, 2h

Other grade(s)
Project report and presentation

Weight: 40/60

Informations complémentaires

TEACHING METHODS

- Interactive lessons: theoretical contributions accompanied by concrete examples and discussions.
- Group case studies: progressive application of concepts to a specific scenario.

NATURE OF MATERIALS
Lesson slides, case studies, data sets

TEACHING INNOVATIONS AND USE OF TECHNOLOGY
All materials available on Moodle. Statistical analysis tools (python, Jamovi...)

PRE-REQUISITES IN TERMS OF KNOWLEDGE AND SKILLS

Knowledge
· Fundamentals of Design Thinking and User-Centered Design
· Basic concepts of Machine Learning (supervised/unsupervised learning, model evaluation)

Skills
· Python for data analysis (pandas, scikit-learn, matplotlib)

Formations dont fait partie ce cours