https://github.com/effinchang/procurement-pattern-detection.

Procurement risk analytics — Neo4j graph patterns (shared addresses, winner rotation) + anomaly detection and data-quality checks
https://github.com/effinchang/procurement-pattern-detection.

anomaly-detection data-quality fraud-analytics graph-database neo4j pattern-detection procurement responsible-ai

Last synced: 25 days ago
JSON representation

Procurement risk analytics — Neo4j graph patterns (shared addresses, winner rotation) + anomaly detection and data-quality checks

• Host: GitHub
• URL: https://github.com/effinchang/procurement-pattern-detection.
• Owner: Effinchang
• License: other
• Created: 2025-09-14T14:28:18.000Z (29 days ago)
• Default Branch: main
• Last Pushed: 2025-09-14T14:49:17.000Z (29 days ago)
• Last Synced: 2025-09-14T16:31:21.561Z (29 days ago)
• Topics: anomaly-detection, data-quality, fraud-analytics, graph-database, neo4j, pattern-detection, procurement, responsible-ai
• Homepage:
• Size: 4.88 KB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE
  • Security: SECURITY.md

Awesome Lists containing this project

README

          
# procurement-pattern-detection

Graph + anomaly detection for procurement integrity (Neo4j + Python) with data quality checks.

**Keywords:** graph database, Neo4j, pattern detection, collusion, anomaly detection, Great Expectations

## Contents

- cypher/collusion_patterns.cql

- notebooks/ (add IsolationForest demo)

- data/nodes.csv, data/edges.csv

## Quickstart

```bash

pip install -r requirements.txt

# quick graph sanity check

python - << "PY"

import pandas as pd, networkx as nx

e = pd.read_csv("data/edges.csv")

G = nx.from_pandas_edgelist(e, "src", "dst", edge_attr=True, create_using=nx.DiGraph())

print(f"Nodes: {G.number_of_nodes()}, Edges: {G.number_of_edges()}")

PY