Real-time-Sentiment-Analysis-on-Innovation-and-Technology

Description

Real-time sentiment analysis on innovation and technology is a valuable tool for businesses to make data-driven decisions based on public opinion. In this project, we developed a tool that extracts and analyzes tweets from Twitter to track public opinion on the latest technology trends. We performed sentiment analysis on the tweets and categorized them as positive, negative, or neutral based on their compound score. Various supervised machine learning classification algorithms like Naive Bayes, Decision Tree, Random Forest, and Logistic Regression were trained on pre-processed data, and their performance was compared based on F1 score, recall, precision, and accuracy. The Logistic Regression model performed the best with an accuracy of 95.02%, indicating that it correctly classified more positive, negative and neutral tweets than the other algorithms. A GUI was developed to display real-time tweet sentiment analysis. Overall, this project provides a tool for businesses to stay up-to-date on the latest technology trends and make informed decisions based on public opinion.