NetraTax
Graph neural network prototype that scores GST invoice fraud risk and shell-company networks.
What it does
Builds a graph of companies (nodes) and GST invoices (edges) from CSV files, trains a graph neural network to label companies as fraudulent or legitimate, and shows risk scores in a Streamlit dashboard and a Flask API. Needs Python 3.9 and PyTorch Geometric; a script makes synthetic data. A SIH 2024 hackathon entry: a prototype needing validation on real GST data, says the README; no licence file.
Licence unknown: we found no licence we could recognise. Check with the maintainers before you reuse the code.
Where it fits on the map
Not on the map: a demo.
The map places maintained projects for a company tax department or a tax authority, without a demo or paid-service flag. Everything else stays on the list with its label.
Similar projects
Of the same kind, for a shared jurisdiction or tax area. Not a ranking.
- amazon-state-sales-and-tax-summaryTurns marketplace order exports into state-by-state sales and tax totals for US sales-tax returns.
- ca-erpnext-zraZRA Smart Invoice VSDC integration for Frappe and ERPNext.
- caseflowTariff classification with legal reasoning for Nigeria's ECOWAS common external tariff and the US HTS.
- CFO StackBeancount accounting run through Claude Code, with jurisdiction packs for Canada, the US, Taiwan and China.For individual taxpayers
- ClasificodeColombian DIAN tariff classification system: RGI rules, pgvector search, LLM reranking and a DIAN scraper.
- conciliador_nfeReconciles Brazilian e-invoices from the tax authority export against ERP entries and lists the gaps.
Maintain this project?
Link to this page from your README, so readers find the facts, the notes and where it fits. The badge is a small SVG served from this site.
[](https://tax-oss.com/projects/netratax/)