Πανεπιστήμιο Μακεδονίας
Revealing evolutionary patterns in dynamic graphs: exploration and visualization techniques
Abstract
dc:descriptionGraphs are a fundamental data structure used to model a wide range of complex systems, such as social networks, biological processes, and transportation systems. These graphs are often dynamic, evolving over time with changes in both their structure and attributes. This introduces significant challenges in the analysis of temporal graphs, which require specialized methods to capture evolving patterns and identify key events. Understanding how systems change over time, and extracting meaningful insights from these changes, is crucial for many real-world applications. This thesis addresses these challenges by developing new techniques for the exploration and visualization of temporal graphs. We introduce the GraphTempo framework, which combines attribute-based and time-based aggregation techniques to enable efficient exploration and analysis of temporal graphs. The framework incorporates novel event detection algorithms that identify key structural and attribute-based changes, such as stability, growth, and shrinkage events, capturing variations in the graph’s structure and attributes. A core aspect of this research is the development of an interactive visualization system that enables users to navigate the temporal evolution of graphs. By combining graph overview, aggregation, and exploration with intuitive visualizations, our approach uncovers trends and highlights important transitions over time. These techniques are evaluated on real-world datasets, including social and collaboration networks, demonstrating their efficiency and effectiveness in revealing insights into dynamic systems. The contributions of this thesis provide novel algorithms and a comprehensive framework for analyzing the evolution of temporal graphs, along with an intuitive visualization system. These techniques offer valuable tools for understanding dynamic systems across various domains, such as social network analysis, epidemiology, and transportation systems, where tracking the evolution of interactions and relationships is crucial. Overall, this research advances the field of temporal graph analysis by offering efficient methods for detecting significant events and visualizing the dynamic behavior of complex systems over time.
Degree
thesis:*- Grantor dc:publisher
- Πανεπιστήμιο Μακεδονίας
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Tsoukanara, Evangelia
- Τσουκανάρα, Ευαγγελία
Subjects
dc:subject × 16- Χρονικά μεταβαλλόμενοι γράφοι
- Γράφοι με χαρακτηριστικά
- Εξερεύνηση γράφων
- Οπτικοποίηση γράφων
- Ανάλυση κοινωνικών δικτύων
- Temporal Graphs
- Attributed graphs
- Graph exploration
- Graph visualization
- Social network analysis
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Φυσικές Επιστήμες
- Επιστήμη ηλεκτρονικών υπολογιστών, θεωρία και μέθοδοι
- Computer and Information Sciences
- Natural Sciences
- Computer science, theory and methods
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/60692