Course Contents
In this lecture, students will get a comprehensive overview of the discipline of "Digital Text Forensics" (DTF) and the various sub-areas that belong to this field of research. A particular focus is on authorship verification (AV), which serves as the basis for several DTF disciplines.
Topics covered include, but are not limited to:
[list]
[*]Preprocessing of texts and their importance in the context of DTF
[*]Representation possibilities for texts (linguistic feature engineering, embedding methods and compression models)
[*]Concepts for comparing texts with respect to stylistic similarity
[*]AV methods and their characteristics
[*]Attacks on DTF methods (aka "authorship obfuscation")
[*]Evaluation concepts and their advantages and disadvantages in the context of DTF
[*]Interpretation/explainability of results and dealing with uncertainty in relation to predictions
[/list]
In addition to discussing the above topics, it is also explained where and, in particular, for what purposes DTF can be used in practice.
Preconditions
Recommended:
Basic knowledge of machine learning, natural language processing, linguistics
Online Offerings
moodle
- Lecturer: HalvaniOren
- Lecturer: WaidnerMichael