Stud.IP Uni Oldenburg
University of Oldenburg
04.03.2024 08:42:10
ThesisTopics

Personal details

Title Development of a Stochastic Method for Detecting Deviant Communication Patterns in Agent-based Energy Systems
Description

Background
In the ongoing digitalization and increasing complexity of energy systems, efficient communication between the involved agents plays a pivotal role. The ability to identify and analyze various communication patterns can be extremely relevant in modeling. The challenge lies in identifying deviant patterns among a multitude of behaviors, especially when they differ from the training data.

Objective of the Thesis
The goal of this thesis is to develop a method capable of recognizing and classifying deviant forms of communication between agents in energy systems. The focus will be, for example, on distinguishing between periodic and event-based communication behaviors. The method should identify new or unexpected patterns in communication behavior that significantly deviate from the patterns represented in the training data.

Home institution Department of Computing Science
Type of work conceptual / theoretical
Type of thesis Bachelor's or Master's degree
Author Malin Radtke, M. Sc.
Status available
Problem statement
  • Conducting a literature review on existing methods of pattern recognition in datasets.
  • Development of a stochastic method for detecting deviant communication patterns. This method should be able to differentiate between known (e.g., periodic) and unknown (e.g., event-based) communication patterns.
  • Analysis and evaluation of the method's effectiveness based on its ability to detect and classify deviant communication patterns.
Requirement
Created 12/02/24

Study data

Departments
  • Energieinformatik
Degree programmes
  • Bachelor's Programme Business Informatics
  • Master's programme Digitalised Energy Systems
  • Master's Programme Computing Science
  • Bachelor's Programme Computing Science
  • Master's Programme Business Informatics
Assigned courses
Contact person