PhiLNet — Phishing Lightweight Network

PhiLNet (Phishing Lightweight Network) is a fusion-based machine learning framework for real-time phishing detection, combining textual analysis and heuristic features to accurately identify phishing websites and messages with low computational overhead.

This repository contains the experimental, training, and evaluation pipeline used to develop PhiLNet. The production-ready model is already deployed as a Chrome extension.

---

📌 Project Overview

PhiLNet is designed to:

• Detect phishing attempts in real time
• Combine text-based features and rule-based heuristics
• Remain lightweight for browser-level deployment
• Support reproducible research and experimentation

⚠️ Note: This repository is not required to use PhiLNet in production. End users can install PhiLNet Vanguard directly from the Chrome Web Store.

---

📂 Repository Structure

After setup, your project directory should look like this:

PhiLNet/
├── data/
│   ├── raw/
│   │   └── dataset/          # Downloaded phishing dataset
│   └── processed/            # Cleaned and feature-engineered data
│
├── models/                   # Trained model artifacts
│
├── notebooks/                # Jupyter notebooks for the pipeline
│   ├── 01_data_collection.ipynb
│   ├── 02_data_preprocessing.ipynb
│   ├── 03_exploratory_data_analysis.ipynb
│   └── 04_model_training_and_evaluation.ipynb
│
├── results/                  # Figures, metrics, and evaluation outputs
│
├── requirements.txt          # Python dependencies
└── setup.txt                 # Environment setup instructions

---

📊 Dataset

PhiLNet uses a publicly available phishing dataset hosted on Mendeley Data.

Download Instructions

1. Download the dataset from: 👉 Phishing Websites Dataset

2. Extract the downloaded archive.

3. Place the entire dataset folder inside:

data/raw/dataset/

The notebooks assume this exact directory structure.

---

🛠️ Requirements

Before running the project, make sure you have:

• Conda (Anaconda or Miniconda)
• Jupyter Notebook
• Python 3.8 (recommended)

Ensure that:

• requirements.txt is located in the base project directory
• You follow the environment instructions exactly as specified in setup.txt

---

⚙️ Setup Instructions

1. Clone the repository

git clone https://github.com/your-username/PhiLNet.git
cd PhiLNet

2. Follow the environment setup

# Read and follow all steps inside:
setup.txt

This typically includes:

• Creating a conda environment
• Installing dependencies from requirements.txt
• Registering the environment as a Jupyter kernel

---

▶️ Running the Project

Once setup is complete, launch Jupyter Notebook:

jupyter notebook

Then run the notebooks in order:

1. 01_data_collection.ipynb

   • Loads and validates the raw phishing dataset

2. 02_data_preprocessing.ipynb

   • Cleans data and extracts textual and heuristic features

3. 03_exploratory_data_analysis.ipynb

   • Analyzes phishing patterns, feature distributions, and correlations

4. 04_model_training_and_evaluation.ipynb

   • Trains the PhiLNet fusion model
   • Evaluates performance and saves results

⚠️ Important: Do not skip notebooks — each step depends on outputs from the previous one.

---

🚀 Deployed Version

PhiLNet is already deployed as a Chrome extension:

🔐 PhiLNet Vanguard

• Real-time phishing detection
• Runs directly in the browser
• Optimized for low latency and minimal resource usage

➡️ Available on the Chrome Web Store: PhiLNet Vanguard

This repository is intended only for research, experimentation, and model development.

---

📄 License & Usage

• This project is intended for academic, research, and educational purposes
• Dataset usage follows the original dataset’s license
• The deployed extension may follow a separate license

---

📬 Contact & Contributions

If you would like to:

• Extend PhiLNet
• Propose improvements
• Reproduce or benchmark results

Feel free to open an issue or submit a pull request.

---

PhiLNet — Built to detect, designed to protect. 🛡️