Whatsapp Chat Analysis

Abstract

This project presents a full-stack WhatsApp chat analysis system designed to extract meaningful insights from exported chat conversations. With the increasing use of WhatsApp in personal, academic, and business communication, analyzing chat data can reveal valuable patterns in user behavior, message activity, sentiment trends, and group interaction. The proposed system accepts WhatsApp chat files in text format and processes them through a modular pipeline using a Streamlit frontend and a Flask backend. The uploaded data is cleaned and structured using regular expressions, Pandas, and Unicode support to handle timestamps, sender names, emojis, media messages, and multilingual text. The system then performs statistical analysis, sentiment scoring, word frequency analysis, emoji tracking, and temporal activity visualization. Advanced features such as topic extraction, activity timelines, response pattern analysis, and interactive charts provide a deeper understanding of conversational dynamics. All processing is performed locally to ensure privacy and security. The application is scalable, easy to use, and suitable for academic research, business communication analysis, and personal chat review. Overall, the system transforms raw WhatsApp chat exports into organized, interactive, and actionable insights through modern data analysis and visualization techniques.