400-1 | Data analysis & machine learning | Computer Science | S7 | ||||||
---|---|---|---|---|---|---|---|---|---|
Lessons : 8 h | TD : 6 h | TP : 6 h | Project : 0 h | Total : 20 h | |||||
Co-ordinator : Christophe ROSENBERGER |
Prerequisite | |
---|---|
Probability and statistics (1A) | |
Course Objectives | |
The student learns algorithms to extract knowledge from an information dataset, to recognize classes of objects (statistical learning) by learning techniques or to solve problems of statistical modeling by non-conventional techniques (evolutionary algorithms). | |
Syllabus | |
- Statistical analysis - Decision tree - Bayesian networks - Neural networks - Machine learning - Genetic algorithms - Boosting |
|
Practical work (TD or TP) | |
Practical work with Matlab Titanic Data Analysis Problem solving with genetic algorithms Learning with a perceptron for the recognition of heart disease |
|
Acquired skills | |
Knowledge of Statistical Learning Knowledge of evolutionary algorithms Skills block: Mobilize the resources of a scientific and technical field related to a specialty -> Level 3: Design adaptive systems based on machine learning, deep learning and data mining. |
|
Bibliography | |
- Gérard Dreyfus , Gérard Dreyfus , Jean-Marc Martinez , Manuel Samuelides "Apprentissage statistique", Edition Eyrolles - Cornuéjols, A and Miclet L.: Apprentissage Artificiel. Concepts et algorithmes (2nd Ed.with revisions and additions - 2006 Eyrolles, 650 p - Christopher M. Bishop, Pattern Recognition and Machine Learning, Springer (2006). |
© 2024 - ENSICAEN ( Legal Notices - Credits )