Internet and classification methods

Garant předmětu:

Lecturer:

Tutor:

Supervisor:

Department of Mathematics

Synopsis:

Attending the course, the students get acquainted with classification methods used in three important internet or general-network applications: spam filtering, recommender systems, and intrusion detection systems. However, they learn more than only how classification is performed when facing these three problems. On the background of the above applications, they get an overall overview about the fundamentals of classification methods. The course is taught in a 2-week cycle, always a 2h lecture and a 2h practice at computer labs.

Requirements:

Syllabus of lectures:

Topic 1. Three important internet applications of classification methods: spam filtering, recommender systems, intrusion detection systems.

Topic 2. Basic concepts concerning classification.

Topic 3. Main kinds of classification methods.

Topic 4. When does a classifier make the least errors on new data?

Topic 5. When is classification comprehensible for a user?

Topic 6. Combining a number of classifiers into a team.

Syllabus of tutorials:

Study Objective:

Study materials:

Recommended reference:

M. Holeňa, P. Pulc, M. Kopp. Classification Methods for Internet Applications. Springer, 2020. ISBN 978-3-030-36961-3.

Kembelec, Chartron, Saleh. Recommender Systems. Wiley 2014.

Pathan. Thes State of the Art in Intrusion Prevention and Detection. CRC Press, 2014.

Stamp. Machine Learning with Applications in Information Security. CRC Press, 2018.

Note:

Further information:

No time-table has been prepared for this course

The course is a part of the following study plans: