📡 Telegram Scraping & Traffic Analysis

A research toolkit for monitoring, scraping, and simulating Telegram channel traffic — built for network traffic analysis and covert communication research.

---

🔍 Overview

This project provides a suite of Python tools to scrape messages from public Telegram channels via web.telegram.org and simulate realistic message traffic for research purposes. It was developed as part of a study on network behavior and traffic fingerprinting in encrypted messaging platforms.

The toolkit includes:

• Two scraping approaches (static HTML parsing and dynamic browser automation)
• A realistic message sender that simulates human-like behavior with mixed media types
• A Jupyter notebook for traffic modeling and analysis

---

📁 Project Structure

Telegram_scraping/
│
├── beautifulsoup.py              # Static scraper using requests + BeautifulSoup
├── selen_scrape.py               # Dynamic scraper using Selenium WebDriver
├── selen_scroll.py               # Selenium helper for auto-scrolling through chat history
├── telegram_txt_sender.py        # Automated message sender with mixed media simulation
├── telegram_text_only_sender.py  # Text-only version of the message sender
├── model 7.ipynb                 # Jupyter notebook: traffic analysis & modeling
└── README.pdf                    # Original documentation

---

⚙️ Features

🕷️ Message Scraping

Module	Method	Dynamic?	Notes
beautifulsoup.py	HTTP + HTML parsing	❌	Lightweight; limited by Telegram's JS rendering
selen_scrape.py	Chrome WebDriver	✅	Handles dynamic content; extracts message ID, sender, timestamp, reply context
selen_scroll.py	Chrome WebDriver	✅	Auto-scrolls to load full chat history

What gets extracted:

• Message ID
• Sender name
• Timestamp (Unix epoch)
• Message text
• Reply context (detects if a message is a reply and extracts the quoted content)

---

📤 Traffic Simulation

telegram_txt_sender.py simulates realistic Telegram traffic by sending randomized messages of varying types at randomized intervals. All activity is logged to a timestamped CSV file.

Message types & probabilities:

Type	Probability	Details
Text	~35.6%	Random alphanumeric strings, exponential length distribution
Image	~58.1%	Random noise images (PNG), size between 2.4 KB – 378 KB
Voice	~6.1%	gTTS-generated audio from random text

Timing: Messages are sent at random intervals drawn from an exponential distribution (mean ~60s, capped at 600s), mimicking realistic human usage patterns.

Output: A CSV file (telegram_messages<timestamp>.csv) with columns:

time, size, message_type

---

🚀 Getting Started

Prerequisites

pip install requests beautifulsoup4 selenium gtts moviepy opencv-python numpy

You'll also need:

• Google Chrome installed
• ChromeDriver matching your Chrome version, added to your system PATH

Setup

1. Clone the repository

   git clone https://github.com/GiladFisher/Telegram_scraping.git
   cd Telegram_scraping

2. Log in to Telegram Web in your Chrome browser before running any Selenium-based scripts. The scripts open web.telegram.org and require an active session.

3. Configure the target channel by editing the url variable in the script you want to run:

   url = 'https://web.telegram.org/k/#@your_channel_name'

---

🧪 Usage

Scrape messages with Selenium

python selen_scrape.py

Launches Chrome, navigates to the target channel, waits for content to load, then extracts and prints all visible messages.

Scrape messages with BeautifulSoup

python beautifulsoup.py

Attempts a static HTTP request to the Telegram Web page. Note: due to JavaScript rendering, this approach has limited effectiveness on modern Telegram Web.

Simulate message traffic

python telegram_txt_sender.py

Opens a Chrome browser pointed at a target Telegram group, then continuously sends randomized messages (text, images, voice) at randomized intervals. Logs all sent messages to a CSV file.

---

📊 Analysis Notebook

model 7.ipynb contains the traffic analysis and modeling work, including:

• Statistical analysis of message size distributions
• Traffic pattern modeling
• Visualization of scraped data

Open it with Jupyter:

jupyter notebook "model 7.ipynb"

---

⚠️ Disclaimer

This project was developed for academic and research purposes only. Please ensure you:

• Only scrape public Telegram channels you have permission to access
• Comply with Telegram's Terms of Service
• Do not use this tool to harass, spam, or violate the privacy of any individual

---

🛠️ Tech Stack

• Python 3.8+
• Selenium — browser automation
• BeautifulSoup4 — HTML parsing
• gTTS — text-to-speech for voice message simulation
• OpenCV — image generation
• MoviePy — video processing
• NumPy — numerical distributions for realistic traffic modeling
• Jupyter Notebook — data analysis

---

👤 Author

Gilad Fisher — GitHub

---

Feel free to open an issue or pull request if you'd like to contribute or report a bug.