GraphVisual — Community Evolution Visualization

Features

🎨

Interactive Visualization

Drag, zoom, and rotate graph nodes using JUNG's graph mouse. Explore community structures visually with cluster-based layouts.

⏱️

Timeline Playback

Animate community graphs across 92 days with play/pause/stop controls. Watch relationships form and dissolve over time.

🏷️

5 Relationship Types

Classify social interactions as friends, classmates, study groups, familiar strangers, or strangers — each color-coded on the graph.

🎛️

Adjustable Thresholds

Tune meeting duration and frequency thresholds per relationship type in real-time to explore different classification parameters.

📊

Network Statistics

Real-time metrics including node/edge counts, graph density, average/max degree, edge weights, isolated nodes, and hub identification.

💾

Export & Annotate

Save visualizations as PNG images, export edge lists for further analysis, and add notes for each timestamp during research.

Relationship Classification

Type	Location	Duration	Frequency
Friends	Public areas	> 10 min	≥ 2/day
Classmates	Classrooms	> 30 min	≥ 1/day
Study Groups	Classrooms	> 20 min	≤ 1/day
Familiar Strangers	Public/paths	< 2 min	> 1/day
Strangers	Public/paths	< 2 min	< 2/day

All thresholds are adjustable at runtime via the Category Panel sliders.

Data Pipeline

📡 Bluetooth Events

Raw Bluetooth proximity data from event_3 table

↓

🔍 findMeetings.java

Extracts meetings — IMEI pairs, start/end time, duration

↓

🔗 matchImei.java

Maps device nodes to IMEI identifiers

↓

📍 addLocation.java

Classifies meeting locations via WiFi access points

↓

🌐 Network.java

Generates edge-list files with parameterized SQL queries

↓

📊 Main.java

Interactive JUNG graph visualization with timeline playback

Getting Started

1 Clone the repository

git clone https://github.com/sauravbhattacharya001/GraphVisual.git
cd GraphVisual

2 Configure database

Set environment variables for PostgreSQL credentials:

export DB_HOST=localhost
export DB_USER=your_user
export DB_PASS=your_pass

3 Build the project

cd Gvisual
ant build

Requires Java JDK 8+ and Apache Ant.

4 Run the pipeline

# Match devices → Extract meetings → Classify locations
java -cp "build/classes:lib/*" app.matchImei
java -cp "build/classes:lib/*" app.findMeetings
java -cp "build/classes:lib/*" app.addLocation

5 Launch the visualizer

java -cp "build/classes:lib/*" gvisual.Main

Tech Stack

☕

Java 8+

Application language

🕸️

JUNG 2.0.1

Graph framework

🖥️

Java Swing

Desktop GUI

🐘

PostgreSQL

Data storage

🐜

Apache Ant

Build system

🧪

JUnit 4

Unit testing

⚙️

GitHub Actions

CI/CD

🔒

CodeQL

Security scanning

Architecture

GraphVisual/
├── Gvisual/
│   ├── src/
│   │   ├── gvisual/
│   │   │   ├── Main.java            # Swing GUI — graph panel, timeline, controls
│   │   │   ├── edge.java            # Edge model (type, vertices, weight)
│   │   │   ├── GraphStats.java      # Network metrics (density, degree, hubs)
│   │   │   └── ShortestPathFinder.java  # Path analysis
│   │   └── app/
│   │       ├── Network.java         # Edge-list generation from DB
│   │       ├── Util.java            # Database connection factory
│   │       ├── findMeetings.java    # Bluetooth → meeting extraction
│   │       ├── addLocation.java     # Meeting location classification
│   │       └── matchImei.java       # Device → IMEI matching
│   ├── test/                        # JUnit test suites
│   ├── lib/                         # JUNG, PostgreSQL JDBC, Commons IO
│   └── images/                      # UI icons
├── docs/                            # This documentation site
├── .github/workflows/               # CI, CodeQL, Pages, Auto-label
└── DATABASE.md                      # Database schema reference

📊 GraphVisual