The aim of this course is to introduce students to selected approaches and methods in the filed of safety and security of information systems, with a special focus on applying natural language processing methods in automatic security analyses.

On completion of this course, students are expected to be able to understand, explain, apply and critically evaluate the introduced approaches and methods in particular contexts of use.

Exploits and prevention of stack overflow attacks; automatic cryptanalysis of substitution ciphers based on statistical language models (n-grams), genetic algorithms and a gradient ascent method; evolutionary approach to automatic analysis of software security flaws based on fuzzy testing, genetic algorithms, context free grammars and Markov chains; automatic classification of unstructured texts based on naive Bayesian classification and logistic regression (i.e., maximum entropy modeling), automatic authorship attribution and malicious code detection based on n-grams.

Lectures introduce the content of this course and present examples and analysis of selected problem scenarios in the field of safety and security of information systems. These are supported by practical exercises in laboratory sessions. In the exercises, students will be encouraged to work individually and in groups to develop systems implementing the methods introduced in lectures.