| license: mit | |
| library_name: sklearn | |
| tags: | |
| - malware-detection | |
| - cybersecurity | |
| - random-forest | |
| - pe-files | |
| - security | |
| datasets: | |
| - dev9269/darkweb-slang-dictionary | |
| # Malware Detector - Random Forest | |
| A Random Forest classifier for PE (Portable Executable) malware detection, trained on the EMBER dataset. | |
| ## Model Details | |
| - **Model Type**: Random Forest Classifier (scikit-learn) | |
| - **Training Data**: EMBER dataset (features from PE file headers, sections, imports, exports, etc.) | |
| - **Framework**: scikit-learn (LightGBM backend) | |
| - **Input**: Extracted PE feature vector (2381 dimensions) | |
| - **Output**: Malicious / Benign classification with confidence score | |
| ## Usage | |
| ```python | |
| import joblib | |
| import numpy as np | |
| model = joblib.load("model.pkl") | |
| features = np.load("sample_features.npy") # 2381-dim feature vector | |
| prediction = model.predict([features])[0] | |
| confidence = model.predict_proba([features])[0] | |
| print(f"Malicious: {bool(prediction)}") | |
| print(f"Confidence: {max(confidence):.2%}") | |
| ``` | |
| ## Performance | |
| | Metric | Score | | |
| |--------|-------| | |
| | Accuracy | ~96% | | |
| | Precision | ~0.95 | | |
| | Recall | ~0.94 | | |
| | F1 Score | ~0.94 | | |
| ## Related Models | |
| - [ember-malware-rf](https://huggingface.co/dev9269/ember-malware-rf) - EMBER-trained variant | |
| - [malware-detector-demo](https://huggingface.co/spaces/dev9269/malware-detector-demo) - Interactive demo Space | |