Published in: Journal of Computer Science and Engineering in Innovations and Research (ISSN No: 3049-1762 online)
Publication Date: June 15, 2025
Due to the enormous user base and real-time platform of Twitter, mining tweets provides insightful views into feelings about various subjects, including politics, product reviews, and social matters. This analysis captures trends in public opinion through classifying tweets. Standard methods include pre-processing tweets to handle language subtleties, hashtags, and emoticons, and applying different types of classifiers. Challenges come from the use of informal language, abbreviations, and sarcasm commonly found in tweets, which decrease model accuracy.This work discuss the sentiment analysis on tweets using BERT,NB and LR.
Sentiment analysis , Accuracy, BERT,NB and LR.
Twitter has emerged as a treasure trove of public sentiment and opinion, where individuals post opinions on different topics ranging from personal life to social, political, and commercial issues. Identification of these sentiments is precious for businesses, researchers, and organizations seeking to discern public mood, forecast trends, and make informed decisions.
Through the use of different algorithms, it is possible to transform raw, unstructured Twitter text data into actionable insights. The project brings together the steps of data collection, pre-processing, training, and validation to create a solid model that can identify the underlying sentiments in tweets. This evaluation finds real-world uses, such as measuring public reaction to events, monitoring brand reputation, and even forecasting market trends. In the end, Twitter Sentiment Analysis serves to convert social media information into valuable data, providing greater insight into public opinion and growing trends.