Machine Learning and Pattern Recognition 1 2015-2016 - IMT4612 - 5 ECTS

Prerequisite(s)

BSc level basics in statistics and mathematics, i.e. expected prior-knowledge in understanding basic statistical methods like descriptive statistics, probability, sampling distributions, and hypothesis testing, as well as basic analysis and matrix algebra.

Expected learning outcomes

Knowledge:

• Understand principles how multidimensional statistical methods differ from one dimensional methods.
• Understand the distribution of information in statistical analysis and meaning in data representation.
• Extract features from raw, measured values of data to be analyzed.
• Program some basic classification and clustering methods and test their validity.
• Program some basic Neural networks methods and test their validity.
• To apply basic statistical and data analysis methods to data relevant in information security, forensics and/or color/media technology.

Skills:

• The students can use relevant scientific methods in independent research and development in machine learning and pattern recognition.
• The students are capable of carrying out an independent limited research or development project in machine learning and pattern recognition under supervision, following the applicable ethical rules.

General competence:

• The students can work independently and are familiar with terminology of machine learning and pattern recognition.

Topic(s)

Main topics are:

• Learning, Intelligence, and Machine learning basics: principles, measures, performance evaluation, method combinations.
• Knowledge representations: discriminant and regression functions, probability distributions, Bayesian classifier.
• Learning as search: Exhaustive search, heuristic search, genetic algorithms.
• Attribute quality measures: measures for classification, measures for regression, application of feature-selection measures.
• Data preprocessing: Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA).
• Supervised symbolic and statistical learning, basics of artificial neural networks.
• Unsupervised Learning and cluster analysis: hierarchical and partial clustering.
• Data classification: Bayesian classifier, k-NN classifier, multi-layered perceptron (MBPN).
• Data clustering: k-means clustering, Self-Organizing map (SOM).
• Classification and clustering validity testing: leave-one-out, ground truth.

• Realize some search methods
• Realize some classification methods
• Realize some clustering methods

Teaching Methods

Lectures
Mandatory assignments
Exercises

Weekly lectures

3 major assignments that include theoretical and practical aspects of the topics (graded)

Weekly exercises that support the lectures - highly recommended in order to prepare for the major assignments (not graded)

Form(s) of Assessment

Written exam, 3 hours
Other

•  Written exam (70%)
•  3 major assignments (30% total, 10% each)
•  The written exam and all the major assignments must be passed

Alphabetical Scale, A(best) – F (fail)

External/internal examiner

Internal and external examiner

Written exam is assessed by external examiner. Assignments are assessed by internal examiner.

Re-sit examination

For the written exam: Re-sit August 2016. The major assignments, if passed, need not be re-submitted.

Examination support

Dictionary is allowed.

Teaching Materials

I. Kononenko, M. Kukar, Machine Learning and Data Mining: Introduction to Principles and Algorithms, Horwood Publishing, Chichester, U.K., 2007, ISBN 1-904275-21-4

• T. Mitchell, Machine Learning, McGraw Hill, 1997.
• R.O.Duda, P.E. Hart, and D.G. Stork: Pattern Classification. 2nd edition., Wiley, 2001.
• S. Theodoridis, and K. Koutroumbas. Pattern Recognition, 3rd edition. Academic Press.

IMT4611