Uncover deep insights from your Telegram chats using various data analysis tools.

Features

• Upload your chat exports to Streamlit-based web interface
• Get basic statistics for your chats and do NLP
• Perform user-level analysis
• Analyze user interactions
• Visualize trends, patterns & geographical data

Initially built for Telegram chat analysis, but also supports WhatsApp exports.

Dependencies

• streamlit
• scikit-learn
• pandas
• scipy
• nltk
• folium
• gensim
• geopy
• googletrans
• pygeohash
• st-annotated-text
• whatstk

Setup

Secrets

To run this app locally, you need to create a secrets.toml file, which should be placed in:

• ~/.streamlit/secrets.toml for macOS/Linux
• %userprofile%/.streamlit/secrets.toml for Windows

Learn more about Streamlit secrets here.

The secrets.toml file should contain your Hugging Face API token:

hf_api_token = "TOKEN"

Follow these instructions to generate it.

Enviroment

Clone the repository:

git clone git@github.com:avrtt/telegram-chat-analyzer.git
cd telegram-chat-analyzer

Install dependencies:

pip install -r requirements.txt

Run:

streamlit run Home.py

Trobleshooting

If you run Streamlit and it says that module is missing, you should probably upgrade Streamlit:

pip install --upgrade streamlit 

Contributing

Feel free to open PRs and issues.

License

Distributed under the MIT License. See LICENSE.txt for more information.