Votre sélection est vide.
Enregistrez les diplômes, parcours ou enseignements de votre choix.
6 crédits
École d'économie de la Sorbonne (EES)
Printemps
École d'économie de la Sorbonne (EES)
Printemps
École d'économie de la Sorbonne (EES)
Printemps
2 crédits
École d'économie de la Sorbonne (EES)
18h
Printemps
Summary: The objective of this course is to show, through the realization of a practical use case, the reality of a project in machine learning in real world and the difficulties encountered. The project will consist of implementing AI explainability methods to counter the black box effect of machine learning methods in a credit risk calculation context. In addition, the students will have to: import data from the Kaggle website, pre-process, implement classical Machine Learning methods and analyze the results succinctly. The next step will be to implement different methods of explicability to best interpret the results obtained. Finally, as all work is intended to be reused, students will have to produce documentation (redaction, git, readme) allowing the reuse of their work. During this course, in addition to the technical aspects, we will quickly address the problems of project management. The main language use for the course is Python.
Course prerequisites: Python programming basics, Machine learning and credit scoring course, Risk Management course
Lecturer: Etienne Gay (Data Science Director - VO2 Group)
Student assessment: Final project in Python
2 crédits
École d'économie de la Sorbonne (EES)
18h
Printemps
2 crédits
École d'économie de la Sorbonne (EES)
18h
Printemps
4 crédits
École d'économie de la Sorbonne (EES)
18h
Printemps
École d'économie de la Sorbonne (EES)
Printemps