Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -1,5 +1,64 @@
|
|
| 1 |
-
---
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: en
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
tags:
|
| 5 |
+
- xgboost
|
| 6 |
+
- machine-learning
|
| 7 |
+
- classification
|
| 8 |
+
- cybersecurity
|
| 9 |
+
- phishing-detection
|
| 10 |
+
datasets:
|
| 11 |
+
- custom
|
| 12 |
+
metrics:
|
| 13 |
+
- accuracy
|
| 14 |
+
- precision
|
| 15 |
+
- recall
|
| 16 |
+
- f1
|
| 17 |
+
---
|
| 18 |
+
|
| 19 |
+
# XGBoost Phishing Detection Models
|
| 20 |
+
|
| 21 |
+
## Model Description
|
| 22 |
+
|
| 23 |
+
XGBoost models trained for phishing detection using URL and HTML content features.
|
| 24 |
+
|
| 25 |
+
This model is trained using XGBoost for binary classification tasks.
|
| 26 |
+
|
| 27 |
+
## Model Architecture
|
| 28 |
+
|
| 29 |
+
- **Model Type**: XGBoost Classifier
|
| 30 |
+
- **Framework**: XGBoost
|
| 31 |
+
- **Task**: Binary Classification
|
| 32 |
+
|
| 33 |
+
## Usage
|
| 34 |
+
|
| 35 |
+
```python
|
| 36 |
+
import joblib
|
| 37 |
+
from huggingface_hub import hf_hub_download
|
| 38 |
+
|
| 39 |
+
# Download the model
|
| 40 |
+
model_path = hf_hub_download(repo_id="th1enq/xgboost_checkpoint", filename="xgboost phishing detection models.joblib")
|
| 41 |
+
|
| 42 |
+
# Load the model
|
| 43 |
+
model = joblib.load(model_path)
|
| 44 |
+
|
| 45 |
+
# Make predictions
|
| 46 |
+
predictions = model.predict(X_test)
|
| 47 |
+
```
|
| 48 |
+
|
| 49 |
+
## Training
|
| 50 |
+
|
| 51 |
+
The model was trained using the XGBoost library with the following approach:
|
| 52 |
+
- Feature extraction from URLs/HTML content
|
| 53 |
+
- Binary classification (legitimate vs phishing)
|
| 54 |
+
- Cross-validation for model evaluation
|
| 55 |
+
|
| 56 |
+
## Files
|
| 57 |
+
|
| 58 |
+
- `xgboost phishing detection models.joblib`: The trained XGBoost model
|
| 59 |
+
- `features.py`: Feature extraction functions
|
| 60 |
+
- `URLFeatureExtraction.py`: URL-specific feature extraction
|
| 61 |
+
|
| 62 |
+
## License
|
| 63 |
+
|
| 64 |
+
This model is released under the Apache 2.0 License.
|