🕵️♂️ Fraud Model Aura (KS-Vijay)
This model uses DistilBERT to classify whether a given grievance or complaint text contains fraudulent intent or behavior. It is trained as part of an intelligent Grievance Redressal Platform to auto-detect fraud-related issues in startup complaints.
🧠 Use Case
Detects if the complaint relates to fraud:
FraudLegitimate
This helps startups or service providers to flag, escalate, or triage suspicious reports quickly.
🔍 Model Summary
- Model Type: Text Classification
- Architecture: DistilBERT (uncased)
- Output Labels:
Fraud,Legitimate - Weights Format:
safetensors - Dataset: Custom (based on complaints.csv)
- Training Framework: PyTorch using 🤗
transformers
📥 Example Input
"I think someone is misusing our company’s KYC information to open fake accounts."
- Downloads last month
- 33
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support
HF Inference deployability: The model has no library tag.
Model tree for KS-Vijay/fraud-model-aura
Base model
distilbert/distilbert-base-uncased