This site uses only proprietary and third party technical cookies. By continuing to browse the site you are agreeing to our use of cookies. I agree I want to find out more
Browse the Department site:
Browse the Teaching site:

Programme of Course "Machine Learning"



Type of course unit:

Master Degree in Computer Science curriculum GSEEM: Elective
Master Degree in Computer Science curriculum NEDAS: Compulsory
Master Degree in Computer Science curriculum SEAS: Elective
Master Degree in Computer Science curriculum UBIDIS: Elective

Level of course unit:

Postgraduate Degrees


1st semester

Number of credits:

Master Degree in Computer Science: 6 (workload 150 hours)


Pasquale Caianiello (PasqualedotCaianielloatunivaqdotit)

1. Course Objectives

Knowledge of main machine learning techniques and their applications to problems of supervised and unsupervised learning.

2. Course Contents and learning outcomes (Dublin Descriptors)

Topics of the course include:

  • Machine learning problems: binary classification, multilabel classification, association, regression, prediction, clusterization.
  • K-nearest neighbor classification and clusterization. Decision trees.
  • Percepton, multilayer perceptron, error backpropagation:theory and practice.
  • Data sets, Training sets, Test sets, learning validation techniques, temporal sequences, learning as model construction, exploration of model space, optimization.
  • Genetic algorithms, ensemble methods and voting strategies, random forest
  • Support Vector Machines

On successful completion of this course, the student should

-know main learning problems and models to solve them; -be able to identify a proper learning model for a given problem; -be able to set up problem data and run learning systems on them; -assess, rate, and measure performance of different learning models solution for a given problem; -discuss and interpret results obtained by learning systems and propose possible improvements; -deliver a final solution system/application for a given learning problem.

3. Course Prerequisites

Knowledge of basic concepts of linear algebra and dicrete mathematics. Ability to develop an experimental or implementation project

4. Teaching methods and language



Reference textbooks

  • Paul Daumé III, A Course in Machine Learning. CIML. 2012.
  • Lutz Hamel, Knowledge Discovery with Support Vector Machines. Wiley. 2009.

5. Assessment Methods

Written test, homeworks, and discussion of an implementation or experimental project

Course information last updated on: 08 settembre 2017, 14:46