MSC-INF102C: Python for data science

Prerequisites

Module MSC-INF 101 E "Introduction to Python & scientific computing"

Objectives

The goal of this module is to give an introduction to various aspects of data science using Python as programming language.

Duration: 21h

Content

Six three-hour sessions of course/practical:

1) We will introduce basic concepts for data analysis, in particular we will focus on objective understanding and data understanding. Important issues we will illustrated from a case study.

2) We will introduce descriptive statistics (univariate and bivariate statistics and visualisation tools). Models, machine learning algorithms and Python for data analysis will be introduced. The case study presented during Lab #1 will be explored.

3) We will introduce modeling techniques and evaluation of models. Linear regression will be applied to the case study discussed during Lab #1 and #2.

4) Linear and polynomial regression, crossed validation.

5) K-means algorithm on messages of a discussion forum.

6) Naive Bayes, sentiment analysis on tweets.

7) Discovery of graphs and centrality measures on graphs using the graph-tool package.

Organization

Examination

A 2-hour written exam without documents and without machine.

Scheduled activities

• C-TP 1 (3h)   Basic concepts of data analysis
• C-TP 2 (3h)   Introduction to descriptive statistics
• C-TP 3 (3h)   Introduction to modeling techniques
• C-TP 4 (3h)   Linear and polynomial regression
• C-TP 5 (3h)   K-means algorithm
• C-TP 6 (3h)   Naive Bayes algorithm
• C-TP 7 (3h)   Centrality measures on graphs

Team