OPCO 2065 - Data analysis with Python - Advanced

Students will work on real-world datasets provided in near-clean form. They will manipulate and clean them if necessary, and the focus will be on applying statistical models. The students will learn how to best choose which methods to apply, and learn the theory behind machine learning and statistical methods such as linear regression, controlled linear regression, logistic regression, classification trees, time series analysis and forecasting. They will learn how to implement a machine learning pipeline (data splitting, model fitting, evaluation) and how to critically conduct experiments and analyze results.

The students are required to be familiar with Python core concepts: data types, control structures, functions, error handling. Knowledge of data handling techniques (opening and manipulating datasets with pandas, data cleaning, basic numpy operations) is also required. They also are required to know basic statistical concepts (summary statistics, simple linear regression).

The materials will be made available on Moodle and GitHub, in the form of presentations and Jupyter Notebook. Typically, one session will be composed of a first theoretical part, followed by hands-on coding with Jupyter notebooks. The students will have to do exercises alone and in small groups, and Wooclap quizzes will also be proposed.

The students will be evaluated continuously during the year with small quizzes during the lectures (e.g. Wooclap quizzes). This evaluation will constitute 20% of the final grade. The remaining 80% will be evaluated through a (group?) project that the students will have to submit at the end of the course. The topic will be decided during the year together with the professors, and will be related to what is studied in the main courses followed by the students. Both the practical results (code, models, data visualization, …) and a final report will be evaluated.

In Class Presence: 2 hours a week every week / 24 hours a semester

Repetition of concepts studied in class: 20 minutes a week / 4 hours a semester

Homework: 1h a week / 12 hours a semester

Research and Preparation for Group Work: 1 hour a week / 12 hours a semester

Research and Writing for Individual Assessment: 20 minutes a week / 4 hours per semester