Comprehensive Sentiment Profiling and URL Mining of WhatsApp Conversations using VADER and NLP Techniques

With over 487.5 million active addicts, WhatsApp is one of the most popular platforms for informal discussion. The point is estimated to take a new communication every 0.5 seconds, denoting it produces a large knob of user- generated content every single day. This paper focuses on the performing sentiment analysis and URL extraction from WhatsApp conversation using Natural Language Processing (NLP) algorithms like VADER (Valence Aware Dictionary and sEntiment Reasoner). The methodology consists of pre-processing conversation data which includes drawing up converse data and organizing the users sentiments into positive, neutral, or negative tags, along with orderly recovery of participated links. URL extraction processes and pattern matching enabled with regular expressions alongside VADER’s preset formulas for scoring sentiments yield precious perceptivity into user exertion, colorful analyses indicated positive connections between content participated externally and the dominating sentiments which redounded in links being participated in the more positive exchanges. Using sentiment pie maps alongside URL frequence graphs allows for farther understanding of the data. These approaches add onto hypotheses concerning stoner actions analytics and can be applied to advancement in covering systems for real- time usages in digital marketing and social media monitoring.