Eevee

Evolving Supply Chain Resilience

Hackathon: ArangoDB Hackathon
Dataset: Amazon Co-Purchase Network

What It Does

Eevee uses graph analytics to evolve supply chains into resilient networks, analyzing the Amazon Co-Purchase dataset with synthetic suppliers and events. It clusters products and suppliers with Louvain detection, highlights risks, and answers queries like “Which suppliers are at risk?” using AI.

Features

• Loads and persists the Amazon dataset in ArangoDB.
• Clusters the network with mock Louvain communities.
• Visualizes risks and communities in an interactive graph.
• Answers supply chain questions with an AI agent.

Setup

1. Clone the Repo:

   git clone https://github.com/JadeSamLee/eevee.git
   cd eevee

2. Install Packages:

   pip install networkx matplotlib plotly arango nx-arangodb langchain langchain-openai

3. Add Credentials:

   • Edit eevee.ipynb with your ArangoDB and OpenAI keys:

     ARANGO_URL = "your-arangodb-url"
     ARANGO_USERNAME = "your-username"
     ARANGO_PASSWORD = "your-password"
     OPENAI_API_KEY = "your-openai-key"

How to Run

1. Execute the Script:

   python eevee.py

   • Downloads the dataset.
   • Creates a sample graph (community_analysis.html).
   • Runs example queries.

2. Sample Queries:

   • "Which suppliers provide Product_1?" → "Nike."
   • "Which suppliers are at risk?" → "Nike and Adidas."

Outputs

• community_analysis.html: Interactive graph with communities and risks.

• Console logs with supplier info.

• ArangoDB for the hackathon platform.

• SNAP for the dataset.