University of Twente Student Theses

Login
As of Friday, 8 August 2025, the current Student Theses repository is no longer available for thesis uploads. A new Student Theses repository will be available starting Friday, 15 August 2025.

Formalization of Tactical Representations in Counter-Strike through Data Preprocessing and Labeling

Kirna, Povilas (2025) Formalization of Tactical Representations in Counter-Strike through Data Preprocessing and Labeling.

[img] PDF
2MB
Abstract:This paper presents a methodology for preprocessing and labelling raw Counter-Strike: Global Offensive (CS:GO) replay files (demos) for tactical analysis. Such analysis is a key part of all kinds of sports; that includes eSports titles. We developed a tool that enables users to label CS:GO demos and automatically predict strategies executed by the terrorist team. This is done by a series of scripts that convert raw data into structured graphs, which are then processed by a Graph Neural Network (GNN) model trained to classify tactics. It makes the job of analysts easier, understanding opponents' tactics more efficient, and enhances learning crucial insights from replays. Our contributions include the first tactically labelled CS:GO demos dataset, a GNN model optimised for tactic prediction, and thus impact eSports analytics in a way that is completely new.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Business & IT BSc (56066)
Link to this item:https://purl.utwente.nl/essays/107405
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page