Data Science and Machine Learning (08_DAT)

• Coefficient : 6
• Hourly Volume: 106.0h (including 72.0h supervised)
  CTD : 9h supervised
  Labo : 63h supervised (and 12h unsupervised)
  Out-of-schedule personal work : 22h
• Including project : 21h supervised and 7h unsupervised project

AATs Lists

Description

1. Introduction to Data Science
   • Definition and scope
   • Overview of data science and machine learning ecosystem
2. Data Exploration
   • Data visualization techniques
   • Descriptive and inferential statistics for understanding datasets
3. Data Preparation
   • Data cleaning and preprocessing
   • Handling missing and categorical data
4. Introduction to Machine Learning
   • Basic concepts and distinction between supervised and unsupervised learning
   • Basic models: Linear regression, logistic regression, and k-means clustering
5. Model Evaluation
   • Cross-validation methods
   • Evaluation metrics for classification and regression models
6. Practical Workshops with Python
   • Using pandas for data exploration and cleaning
   • Introduction to scikit-learn for building and evaluating simple models

Learning Outcomes (AAv)

• AAv1 [heures: 0, B2, D3]: By the end of semester S8, students will be able to apply a machine learning model, adapt it to a given context, implement it with a Python machine learning framework (scikit-learn or PyTorch), and evaluate its performance using relevant metrics.

• AAv2 [heures: 0, B3, D3]: By the end of semester S8, given a dataset, students will be able to choose an appropriate machine learning model to solve a regression or classification problem and argue their choice based on performance comparison.

• AAv3 [heures: 0, B2, B3, D3]: By the end of semester S8, students will be able to use data preprocessing techniques to improve machine learning model performance. They will be able to justify their choice by presenting performance comparisons.