Update README.md
Browse files
README.md
CHANGED
|
@@ -33,7 +33,7 @@ Fine-tuned DistilBERT model for Personal Identifiable Information (PII) detectio
|
|
| 33 |
- **Task**: Token Classification (Named Entity Recognition)
|
| 34 |
- **Languages**: English
|
| 35 |
- **License**: MIT
|
| 36 |
-
- **Fine-tuned on**: AI4Privacy PII-
|
| 37 |
|
| 38 |
## Supported PII Entity Types
|
| 39 |
|
|
@@ -62,7 +62,6 @@ This model can detect 56 different types of PII entities including:
|
|
| 62 |
**Professional Information:**
|
| 63 |
- JOBTITLE, JOBTYPE, JOBAREA, COMPANYNAME
|
| 64 |
|
| 65 |
-
**And many more...**
|
| 66 |
|
| 67 |
## Usage
|
| 68 |
|
|
@@ -115,24 +114,3 @@ This model is designed for:
|
|
| 115 |
- Performance may vary on very short or very long texts
|
| 116 |
- Should be validated on your specific use case
|
| 117 |
|
| 118 |
-
## Ethical Considerations
|
| 119 |
-
|
| 120 |
-
This model is intended to help protect privacy by identifying PII. Users should:
|
| 121 |
-
- Test thoroughly on their specific data
|
| 122 |
-
- Implement appropriate safeguards
|
| 123 |
-
- Consider the legal requirements in their jurisdiction
|
| 124 |
-
- Be aware that no automated system is 100% accurate
|
| 125 |
-
|
| 126 |
-
## Citation
|
| 127 |
-
|
| 128 |
-
If you use this model, please cite:
|
| 129 |
-
|
| 130 |
-
```bibtex
|
| 131 |
-
@misc{bert-pii-detection,
|
| 132 |
-
title={BERT PII Detection Model},
|
| 133 |
-
author={SoelMgd},
|
| 134 |
-
year={2025},
|
| 135 |
-
publisher={Hugging Face},
|
| 136 |
-
url={https://huggingface.co/SoelMgd/bert-pii-detection}
|
| 137 |
-
}
|
| 138 |
-
```
|
|
|
|
| 33 |
- **Task**: Token Classification (Named Entity Recognition)
|
| 34 |
- **Languages**: English
|
| 35 |
- **License**: MIT
|
| 36 |
+
- **Fine-tuned on**: AI4Privacy PII-42k dataset
|
| 37 |
|
| 38 |
## Supported PII Entity Types
|
| 39 |
|
|
|
|
| 62 |
**Professional Information:**
|
| 63 |
- JOBTITLE, JOBTYPE, JOBAREA, COMPANYNAME
|
| 64 |
|
|
|
|
| 65 |
|
| 66 |
## Usage
|
| 67 |
|
|
|
|
| 114 |
- Performance may vary on very short or very long texts
|
| 115 |
- Should be validated on your specific use case
|
| 116 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|