GraphVizScript

GraphVizScript — is a desktop application with a graphical interface designed to visualize graphs, analyze their structure using modern and classical algorithms, and save the analysis results to various storages.

Screenshots:

Dark theme - 10k vertices and 20k edges:

White theme:

Dark Theme - 5k vertices and 10k edges:

Video demonstration on YouTube:

🔧 Main functionality

• Graph analysis:
  • Algorithm for graph layout on the plane
  • Search for communities using the Louvain algorithm
  • Classical algorithms:
    • Extraction of strongly connected components
    • Search for cycles from a given vertex
    • Construction of a minimum spanning tree (MST)
    • Search for the shortest path - Ford Bellman
• Saving and loading graphs:
  • Storing graphs in SQLite
  • Integration with Neo4j

🖼 Interface

The application uses the MVVM architectural pattern. Graphs are displayed in a convenient interactive form with visualization of the algorithms' results

🧪 Testing

• Unit tests for algorithms, SQLite
• Integration tests with user scenarios (SQLite saving and loading + Ford Bellman and Find Cycles)

📦 Technologies

• Language: Kotlin
• Async: Kotlin Coroutines
• GUI: Jetpack Compose
• DataBases: SQLite, Neo4j
• Building system: Gradle
• Testing: JUnit5

🚀 Get started

• To download project:

    git clone git@github.com:spbu-coding-2024/trees-trees-team-4.git

• To build project:

    ./gradlew build

• To run application:

    ./gradlew run

• To run a local instance of Neo4j, make sure Docker is installed on your system. Then, start a Neo4j session using the following command:

 docker run -d --name neo4j -p 7474:7474 -p 7687:7687 -e NEO4J_AUTH=neo4j/Test1234554321 -v /home/vkinzin/data/neo4j/data neo4j

This will run Neo4j locally at http://localhost:7474 with default credentials:

Username: neo4j
Password: test

---

Model

Our model supports multiple graph types:

• Directed Graph
• Undirected Graph
• Directed Weighted Graph
• Undirected Weighted Graph

Algorithms

The application provides a collection of classical graph algorithms, implemented in the algorithms package and available through the user interface:

• MST algorithm — Kraskal finds a minimum spanning forest of an undirected edge-weighted graph. If the graph is connected, it finds a minimum spanning tree

• Ford-Bellman algorithm — finds shortest paths in graphs that may include negative edge weights

• Find Cycle algorithm — determines whether a cycle exists involving a specified vertex

• Strongly connected - Kosaraju's algorithm — identifies and extracts strongly connected components in directed graphs

In addition to classical algorithms, the application features built-in layout algorithms for automatic graph visualization:

• Find Communities - Louvain Algoritms — The Louvain method for community detection is a greedy optimization method intended to extract non-overlapping communities from large networks

• Graph Layout - ForceAtlas2 is a force-directed layout algorithm based on physical simulation principles like gravity, repulsion, attraction

All algorithms are modularly implemented, making them reusable outside the user interface context

Supported languages:

• English
• Russian
• Chinese

📄 License

This project is licensed under the MIT License.

📌 Authors

• Vladimir Pugovkin'
• Rafael
• Tatyana Gromova

© 2025