RazanMujawar/Malware-Detection-and-Classification
GitHub: RazanMujawar/Malware-Detection-and-Classification
基于随机森林算法与 PE 头静态分析的机器学习恶意软件检测系统，提供 Flask Web 界面实现对 Windows 可执行文件的良性与恶意分类。
Stars: 0 | Forks: 0
# 恶意软件检测与分类
## 📌 项目简介
本项目旨在利用**机器学习**技术检测并分类 Windows 可执行文件（`.exe`）。通过**静态分析**和**基于模式的学习**，系统能够识别文件是**恶意软件**（Malware）还是**良性文件**（Benign）。
该工具采用 Python 作为后端，并配备了一个简洁的 Web UI，展示了在网络安全实时分析中的实用方法。它作为一个基础模型，可以扩展为企业级安全工具，或与防病毒软件及威胁情报系统进行集成。
## 🧪 主界面
提供了一个简洁、极简的 UI，用户可以在此上传 `.exe` 文件以进行恶意软件检查。
## 🚨 预测结果
扫描完成后，系统会立即返回结果：
- **MALWARE**（如果检测到威胁）
- **BENIGN**（如果文件安全）
## 🚨 恶意软件 (Malware)
##✅ 良性文件 (Benign)
### ✅ 使用 Conda（推荐）
```
# 步骤 1：创建虚拟环境
conda create -n malware_detector python=3.10
# 步骤 2：激活环境
conda activate malware_detector
# 步骤 3：Clone repository
git clone https://github.com/dhirajrhatwal8788-cell/Malware-Detection-and-Classification.git
cd Malware-Detection-And-Classification
# 步骤 4：安装 dependencies
pip install -r requirements.txt```
----
📦 Python Dependencies
These are the primary packages used in the project:
| Package | Purpose |
| -------------- | -------------------------------------------- |
| `Flask` | Backend web framework |
| `scikit-learn` | ML model (Random Forest) |
| `pandas` | Data handling and manipulation |
| `numpy` | Numerical computation |
| `joblib` | Saving and loading ML models |
| `pefile` | Extracting static features from `.exe` files |
| `werkzeug` | Secure filename handling in Flask |
Install them all using:
pip install flask scikit-learn pandas numpy joblib pefile werkzeug
---
🗂️ Directory Structure
Malware-Detection-System/
├── app/ # Core application logic
│ ├── feature_extractor.py # Feature extraction from .exe files
│ └── predict_malware.py # Prediction function / wrapper
│
├── scripts/ # Scripts used for training or testing
│ ├── train_model.py # Model training script (if applicable)
│ └── create_dataset.py # Dataset preparation script
│
├── models/ # Stored trained models
│ ├── classifier.pkl
│ └── malware_model.pkl
│
├── data/ # Input datasets for training/testing
│ ├── dataset.csv
│ ├── PE_Dataset_Label.csv
│ └── sample_dataset.csv
│
├── templates/ # HTML templates for Flask UI
│ └── index.html
│
├── uploads/ # Temporary uploaded .exe files
│ ├── fake_malware.exe
│ └── notepad.exe
│
├── static/ # (Optional) for future CSS or JS
│
├── app.py # Main Flask server
├── requirements.txt # List of Python packages
├── README.md # Project documentation
----
🚀 Running the Application
To start the web app locally:
python app.py
Then visit the application in your browser at:
http://127.0.0.1:5000/
----
🛠️ Tech Stack
Layer Tools Used
Frontend HTML, CSS
Backend Python, Flask
ML Framework scikit-learn (Random Forest)
File Parsing pefile (Python PE Parser)
Hosting Localhost (Flask Server)
---
🚧 Planned Enhancements
Add threat intelligence APIs (VirusTotal, Hybrid Analysis)
Extend support for ELF and Mach-O binaries (Linux/macOS)
Deploy with Docker for better security
Add advanced visualization for feature analysis
Build an authentication module for multi-user access
```
标签：Apex, DNS 反向解析, Flask, Python, 云安全监控, 后端开发, 多模态安全, 无后门, 机器学习, 逆向工具, 静态分析