Code
MGYF INF 4602
Level
M1
Field
Informatique
Language
Français/French
ECTS Credits
3
Class hours
14
Total student load
40
Program Manager(s)
Department
- Data analytics, Économie et Finances
Educational team
Introduction to the module
This module offers a practical introduction to Python programming applied to data analysis and artificial intelligence. Following a "learning by doing" approach, it equips apprentices with hands-on tools to manipulate real data and understand the fundamental mechanisms of AI models — directly connected to their professional context in work-study programmes.
Sessions combine short theoretical inputs with practical workshops in Jupyter Lab. Apprentices are encouraged to draw on data and challenges encountered in their host company. The course also addresses digital responsibility: algorithmic bias and AI ethics.
Learning goals/Programme objectives
- 1. S’approprier les usages avancés et spécialisés des outils de l’intelligence digitale en s’assurant de leur impact durable et responsable
Objectifs d'apprentissage
- 1.1 - Audit advanced and specialised uses of digital intelligence tools in order to deploy them appropriately, taking into account the strategic context of organisations.
- 1.2 - Use digital intelligence tools efficiently to support the societal, digital, energy and environmental transformations of organisations, ensuring their sustainable and responsible impact.
Rubrics
- Write and run Python scripts for data processing (loading, cleaning, aggregating, visualizing)
- Use the Pandas library to manipulate real DataFrames
- Produce clear and meaningful visualizations from data
- Understand the fundamentals of a machine learning model (regression, classification)
- Identify and question potential biases in a dataset or AI algorithm
- Connect a business problem to a data analysis and AI approach
*******************************************************************
- Écrire et exécuter des scripts Python de traitement de données (chargement, nettoyage, agrégation, visualisation)
- Utiliser la bibliothèque Pandas pour manipuler des DataFrames réels
- Produire des visualisations lisibles et argumentées à partir de données
- Comprendre les principes fondamentaux d'un modèle de Machine Learning (régression, classification)
- Identifier et questionner des biais potentiels dans un jeu de données ou un algorithme d'IA
- Connecter une problématique métier à une démarche d'analyse de données et d'IA
Content : structure and schedule
Session 1 (2h) — Python fundamentals: Jupyter Lab; variables, types, operators; control structures (if/else, loops). Lab: first scripts on a simple dataset.
Session 2 (2h) — Functions and data structures: function definition; lists and dictionaries. Lab: structured data processing with functions.
Session 3 (2h) — Pandas: data analysis: DataFrame creation, CSV/Excel import; selection, filtering, sorting; descriptive statistics. Lab: sector-specific dataset exploration.
Session 4 (2h) — Visualisation: matplotlib, seaborn (histograms, barplots, scatterplots); principles of effective visualisation. Lab: commented charts on a real dataset.
Session 5 (2h) — Introduction to AI and Machine Learning: AI overview (symbolic AI, ML, deep learning); supervised learning: linear regression with scikit-learn; train/test/evaluate pipeline. Lab: build a first predictive model.
Session 6 (2h) — Classification and algorithmic bias: decision tree or logistic regression; algorithmic bias and AI ethics; AI use cases in business. Lab: detect bias in a recruitment or credit dataset.
Session 7 (2h) — Business application + Assessment: AI case study linked to the apprenticeship sector; model limits (overfitting, interpretability, ethics); oral presentation of individual reports (10 min/student).
Sustainable Development Goals
SDG 4 — Quality Education: This course develops technical and professional skills that are directly transferable to the workplace. The “learning by doing” approach and the program's integration into a work-study context enhance employability and access to skilled careers in data and AI.
SDG 9 — Industry, Innovation, and Infrastructure: Apprentices acquire the tools of digital transformation (Python, Pandas, AI, scikit-learn) that are at the heart of industrial innovation and organizational modernization.
SDG 10 — Reduced Inequalities: Session 6 addresses algorithmic biases and statistical discrimination in AI models. Apprentices learn to identify these biases and to question the accountability of automated decision-making systems.
**********************************************************************************
ODD 4 — Éducation de qualité : Ce cours développe des compétences techniques professionnelles directement transférables en entreprise. L'approche "learning by doing" et l'ancrage dans le contexte de l'alternance renforcent l'employabilité et l'accès à des métiers qualifiés liés à la donnée et à l'IA.
ODD 9 — Industrie, innovation et infrastructure : Les apprentis acquièrent les outils de la transformation numérique (Python, Pandas, IA, scikit-learn) qui sont au cœur de l'innovation industrielle et de la modernisation des organisations.
ODD 10 — Réduction des inégalités : La séance 6 traite des biais algorithmiques et de la discrimination statistique dans les modèles d'IA. Les apprentis apprennent à identifier ces biais et à questionner la responsabilité des systèmes automatisés de décision.
Number of SDG's addressed among the 17
3
Learning delivery
synchrone
Pedagogical methods
- Apprentissage par la pratique ("learning by doing") : chaque concept est immédiatement appliqué sur un dataset réel
- Ateliers individuels sur Jupyter Lab (ordinateur portable requis)
- Mise en situation professionnelle : les apprentis sont invités à mobiliser des données ou problématiques issues de leur entreprise d'accueil
- Correction collective et discussions en groupe sur les approches utilisées
- Rapport individuel d'analyse ancré dans le contexte de l'alternance, présenté à l'oral
Evaluation and grading system and catch up exams
CF1 — Regular session:
- Individual report: Python analysis of a real dataset related to the apprenticeship context (Jupyter notebook submitted via Moodle) — 50%
- Oral presentation: defence of the report during session 7 (10 min + Q&A) — 50%
An individual written exam (1h, Python exercises on provided data, open-book) may be added at the teacher's discretion, communicated at the start of the course with updated weighting.
CF2 — Resit:
Individual report on a new dataset provided by the teacher (Jupyter notebook submitted within a communicated deadline), followed by an oral defence (10 min + Q&A). Replaces all CF1 grades.
Module Policies
Teacher–Student Communication: The teacher will contact students via their IMT-BS/TSP institutional email and Moodle. Students are responsible for regularly checking their institutional mailbox.
Attendance: Attendance is mandatory. Practical workshops cannot be recovered remotely.
Equipment: Each student must have a laptop with Python 3 and Jupyter Lab installed (or access to JupyterHub). Setup will be verified in session 1.
Academic integrity: Individual work must be completed independently. Use of generative AI tools is allowed if declared and commented. Students must be able to explain their submitted work orally.
Textbook Required and Suggested Readings
Obligatoire (en ligne, gratuit) :
- McKinney, W. Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython (3e éd., O'Reilly). Chapitres 5-10 pour Pandas.
- Documentation officielle Pandas : https://pandas.pydata.org/docs/
- Documentation scikit-learn : https://scikit-learn.org/stable/getting_started.html
Suggérées :
- Rajagopalan, G. A Python Data Analyst's Toolkit (Apress)
- Vincent & Le Goff. Apprenez à programmer en Python (OpenClassrooms, gratuit en ligne)
- Stephenson, B. The Python Workbook (Springer)
Pour aller plus loin :
- Introduction Python interactive : https://nbviewer.jupyter.org/github/phelps-sg/python-bigdata/blob/master/src/main/ipynb/intro-python.ipynb
- Maîtriser Jupyter : https://realpython.com/jupyter-notebook-introduction/
Keywords
Python, Pandas, artificial intelligence, machine learning, scikit-learn, Jupyter, data visualization, algorithmic bias, work-study program, digital transformation **** Python, Pandas, intelligence artificielle, machine learning, scikit-learn, Jupyter, visualisation, biais algorithmiques, alternance, transformation numérique
Prerequisites
Aucun prérequis en programmation. Une aisance avec les outils bureautiques (Excel, tableurs) et une familiarité avec la notion de tableau de données est un plus. Les étudiants doivent disposer d'un ordinateur portable fonctionnel.