| | --- |
| | license: apache-2.0 |
| | datasets: |
| | - CiferAI/cifer-fraud-detection-mini-dataset |
| | language: |
| | - en |
| | metrics: |
| | - accuracy |
| | --- |
| | |
| | # 🧠 Cifer Fraud Detection Mini Model |
| | `(cifer-fraud-detection-mini-model)` |
| |
|
| | ## 🧾 Overview |
| | This is a compact binary classifier trained to detect fraudulent transactions using the **Cifer Mini Fraud Detection Dataset**. It serves as an example model for demonstrating fully encrypted training using Fully Homomorphic Encryption (FHE). |
| |
|
| | The model architecture is lightweight and optimized for short training cycles, making it ideal for testing encryption workflows, verifying pipeline correctness, or onboarding new users. While not production-scale, it reflects the same privacy-first principles as larger models in the Cifer ecosystem: decentralized learning, secure computation, and zero raw data exposure. |
