Introduction

This repository hosts the code for the paper:
GeneticPrism: Multifaceted Visualization of Citation-based Scholarly Research Evolution

🔗 Online System: https://genetic-flow.com / https://geneticflow.ye-sun.com/

🎞️ Demo Video: https://youtu.be/zVbM7lgA6Ig

See User Manual and Appendix in Wiki pages.

Abstract

Understanding the evolution of scholarly research is essential for academic decision-making (e.g., research planning and frontier exploration). Existing platforms like Google Scholar rely on abstract numerical indicators lacking contextual depth, while visualization approaches rarely leverage curated self-citation data to depict individual scholars’ evolution.

This work introduces:

A 3D prism metaphor visualizing scholars’ research profiles
A scroll metaphor visualizing structured topic evolution via streamgraphs and inter-topic flow maps
Six-degree-impact glyphs highlighting interdisciplinary breakthroughs
Evaluations through case studies (Turing Award laureates, visualization venues) and user studies

Dataset

Processed from the open-source Academic Graph:

v1 (up to Sept. 2022): process Microsoft Academic Graph (MAG) to construct GF Graph (from KDD’23 paper, github repo)
v2 (up to Oct. 2024): MAG fused with OpenAlex

🔗 Download v2 dataset: Hugging Face. Due to the dataset's size, it is divided into two compressed archives.

The csv.tar.gz contains CSV files covering all research fields except Artificial Intelligence (AI). After extraction, place these CSV files directly in your project root directory.
The AI.tar.gz contains only AI-related data – extract its CSV files into the project's csv/.

The system remains fully functional if only one archive (either AI or Non-AI) is installed, enabling flexible data management based on research needs.

Deployment

Step 1: Data Preparation

Place the extracted CSV files in the project root directory. The directory structure should look like this:

GeneticPrism/
├── csv/
│   ├── AI/  # Contains AI-related data
│   │   ├── links/
│   │   ├── papers/
│   │   ├── paperIDDistribution.csv
│   │   ├── top_field_authors.csv
│   │   └── field_leaves.csv
│   └── <field>     # Contains other research fields
│       ├── links/
│       ├── papers/
│       └── ...
├── manage.py
└── ...

Step 2: Install Dependencies

conda create -n GFVis python=3.11
conda activate GFVis
pip install -r requirements.txt

Step 3: Run the Server

Option A: Direct run

python manage.py runserver 0.0.0.0:9001

Option B: Background run (persistent)

nohup python manage.py runserver 0.0.0.0:9001 2>&1 &

Access the system at: http://<your-ip>:9001
Use ctrl + c to terminate direct runs
Monitor background processes via tail -f nohup.out