خيارات التسجيل
Data Analysis (Analyse de données): Master RSD, Semester 1, UEM 1.1
This course introduces students to the full process of analysing data, from raw datasets to interpretable results, using Python. It starts with the data analysis life cycle (KDD, CRISP-DM) and a review of the mathematical and univariate statistical tools the course relies on. Students then learn to prepare real data by handling missing values, outliers, and feature transformations. The course moves on to relationships between variables, covering correlation, the chi-square test, ANOVA, and linear regression. It then presents factorial methods for reducing and visualising multidimensional data: PCA, CA, MCA, and FAMD. It ends with clustering techniques, namely hierarchical clustering and K-Means, together with their validation using inertia and the silhouette score. Each concept is illustrated with worked examples on public datasets, and the lab sessions put the methods into practice with pandas, NumPy, SciPy, statsmodels, scikit-learn, Prince, and seaborn.