🤖 Detecting Twitter Bots using Machine Learning

📌 Project Overview

A machine learning-based system to identify and flag malicious Twitter bots with 74% accuracy.
The model analyzes user behavior, account metadata, tweet frequency, and URL patterns to differentiate between human users and bots.

---

🚀 Key Features

• ✅ Real-time and offline Twitter data analysis
• ✅ Feature extraction: Tweet frequency, followers/following ratio, account metadata
• ✅ Logistic Regression classifier with 74% detection accuracy
• ✅ GUI built using Tkinter for user-friendly interaction
• ✅ Malicious URL detection module

---

🛠️ Tech Stack

• Languages: Python
• Libraries: Pandas, NumPy, Scikit-Learn, Matplotlib, Seaborn, Tkinter
• Tools: Jupyter Notebook, GitHub, Google Colab

---

📂 Project Structure

detecting-twitter-bots-ml/
│── data/                 # Dataset files
│── docs/                 # Screenshots & visualizations
│── notebooks/            # Jupyter Notebooks
│── src/                  # Source code (model scripts)
│── requirements.txt      # Dependencies
│── README.md             # Project documentation
│── LICENSE               # MIT License

---

📊 Model Performance

• Logistic Regression Accuracy: 74%
• Evaluated using Confusion Matrix, Precision, Recall, and ROC Curve.

---

🚀 How to Run

1️⃣ Clone the repository:

git clone https://github.com/mohammed-imad-umar/detecting-twitter-bots-ml.git

2️⃣ Install dependencies:

pip install -r requirements.txt

3️⃣ Run the project:

• Open src/Main.ipynb in Jupyter or Colab
• OR double-click src/run.bat for GUI version

---

📸 Screenshots & Visualizations

🔹 GUI & Output Screens:

🔹 Performance Graphs:

---

📫 Contact

⭐ "Detecting fake accounts, one bot at a time!"