Basic notions in applied probabilities theory and mathematic optimization.
Be able to perform a full analysis of a database, as in: prepare the database for the analysis, perform descriptive statistics, understand the importance of separating data, know how to identify problems (regression versus classification) from the variable to study, know how to set up a regression/classification algorithm by choosing the appropriate Python library, know how to analyse the output of a supervised learning algorithm, know how to evaluate the quality and predictive power of a model, know how to compare one or more different models (with many different types of parameters), know how to identify improvement possibilities of a model.
This course is about learning artificial intelligence (& Data science) and the Python programming language. It has 2 main parts: The first part is a lecture containing some theoretical exercises (treated during the course in the form of interactive tutorials) and practical work on a machine (aimed at learning the language of
Python programming). This course, after an initiation phase and/or Python programming refresher, starts with a quick introduction to Syllabus cycle Master BDD Course 11 Data Science with reminders of probability and statistics.
Then, the course focuses on supervised learning algorithms in presenting in detail the simple and multiple linear regression before
to attack the classification algorithms. Classification algorithms binary such as the “k nearest neighbours”, logistic regression or even
the Naïve Bayes are presented or recalled. The Perceptron algorithm is presented after a more formal introduction to artificial intelligence. The
concepts of deep learning and error backpropagation are quickly mentioned throughout the presentation of the Perceptron multilayer. Multi-class classification algorithms and strategies are also presented. For each of these learning algorithms supervised, at least one illustration example on a dataset is presented in Python. Finally, a summary of good practices for the analysis complete database is presented.
The second part of the course focuses on a Project carried out in the form of TP (with a database and a specialty-oriented topic for each
Option or groups of Options). The project, which covers all major aspects of the course, implements theoretical and practical skills taught in the first part.
Report and Practical work assessment.
