Objectives

Today, enterprises collect troves of data about their clients: historical purchases, responses to marketing events, web search logs, etc. In today’s data-driven economy, data can assist us in better understanding our customers, and in taking more informed decisions about our business.

Our goal in this class is to understand the basic terminology of data science and machine learning (regression, classification, visualization, text analytics, recommender systems, etc), comprehend the potential pitfalls, get a general understanding of how to address real-world problems using Python code.

Contents

Some topics that we will cover in the course include:

• Introduction: Data Mining and Machine Learning, Concepts and Terminology. Applications: Targeted Marketing, and Customer Modeling
• Data Preparation and cleaning for knowledge discovery
• Exploratory Data Analysis and Data Visualization
• Predicting numerical values with Linear Regression
• Predicting categorical values. Classification. Decision Trees, Nearest Neighbor Classification, Logistic Regression
• Evaluation of a predictive model
• Recommender Systems
• Text Analytics
• Neural Networks

References

These are recommended but not required textbooks.

- Data Science for Business: What You Need to Know about Data Mining and Data-Analytic Thinking, Foster Provost, Tom Fawcett, ISBN-13: 978-1449361327

- Data Mining: Practical Machine Learning Tools and Techniques, Ian H. Witten, Eibe Frank, Second Edition, 2005, ISBN: 0-12-088407-0

Pre-requisites

- Good knowledge of Python and object-oriented-programming (OOP) in Python

Evaluation

First attempt

Exam:

Written 2h00 hours

Documentation:

Allowed with restrictions

Calculator:

Allowed with restrictions

Evaluation:

The written exam is an ENEP (examen numérique en présentiel). The grade for this course is calculated as follows:

• Programming assignments in Python: 30%
• Group project in Python: 30%
• Written exam (ENEP): 40%

The written exam consists of multiple choice and open questions. Restricted documentation: 1 page with notes (double-sided).

For this course, class participation is important. You are expected to share your thoughts, help your colleagues and participate in the discussions in class and on the class forums.

Retake

Exam:

Written 2h00 hours

Documentation:

Allowed with restrictions

Calculator:

Allowed with restrictions

Evaluation:

The written exam is an ENEP (examen numérique en présentiel). The grade is calculated as:

Examen intégratif (ENEP): 100%

The written exam consists of multiple choice and open questions. Restricted documentation: 1 page with notes (double-sided).