Instagram Chat Analyzer - UVEC 2025 HACKATHON

A fully self-hosted web-application, displays in a visual manners a peek into the users digital life. As people spend more and more time online, they wish to know more about how this time is spent. We solve this through informing the user on

• Top friends (by message count)
• Activity Timeline
• Word Frequenecy
• Emoji Insights
• Message Patterns
  • Most active time of day for recieveing and sending
  • Weekday vs Weekend heat maps
  • Average message length by friend
  • Conversation Streaks

Finally, wrap all of this information in a summary card with

• DM Wrapped
  • Your longest conversation streak
  • Total messages sent
  • Total messages recieved
  • Most Used Word
    Giving the user the ability to reflect on the conversations they spend time on.

Quick Start

Install our dependencies, by

python -m venv venv
source venv/bin/activate
pip3 -r requirements.txt

Flask App:

cd dev
python3 app.py

NextJs App

cd client
npm i react-d3-cloud --legacy-peer-deps
npm i

npm run dev

Then go to instagram webiste and [ADD HERE LATER]

Tech Stack

We will be using a NextJS for the front end and Flask - python for the backend. Our main dataformat is the pandas DataFrame.

Division Of Labour

This project has three components: Front End, Back End Infra, Analysis Logic.
Ali
Worked on the Front-End
Ngan
Worked on the Flask End-Points and Business Logic

Khanh
Worked on Analysis Logic
Abil
Worked on Analysis Logic