jamal-ibrahim/risk_assesment

Model Repo: jamal-ibrahim/risk_assesment

🛡️ Risk Assessment Model Model Repo: jamal-ibrahim/risk_assesment

This is a text classification model fine-tuned on top of DistilBERT to assess reputational risk levels in textual reviews or comments. It predicts whether the input text represents:

Low risk

Medium risk

High risk

🚀 Model Details Base model: DistilBERT

Fine-tuning data: ~1,000 entries of reviews, manually classified into risk levels

Intended use: Automated screening of reviews, comments, or user feedback for reputational risk management

Language: English

💻 Usage You can load the model using the Hugging Face pipeline API:

"""""""""""""""""" from transformers import pipeline

pipe = pipeline("text-classification", model="jamal-ibrahim/risk_assesment")

result = pipe("The company has been accused of fraud.") print(result) """""""""""""""""""""" Example output:

[{'label': 'High', 'score': 0.92}]

📝 Classes Label Description Low No significant reputational risk Medium Possible moderate concerns High Severe reputational implications

⚠️ Limitations & Considerations Trained on a small, domain-specific dataset

Not suitable for legal or financial decision-making without human oversight

May not generalize perfectly to other text domains

Detailed training parameters and metrics are not available anymore