Machine Learning Techniques For Twitter Data Using Sentimental Analysis

In current era, the expressive growth of social media platforms like Twitter has made it an invaluable source for understanding public opinion and sentiment. This paper discovers various machine learning techniques applied to twitter data for sentiment analysis. The study also investigates into the preprocessing steps essential for Twitter data, such as tokenization, stop-word removal, handling emojis and dealing with slang. The results suggest that machine learning models, especially those utilizing word embeddings and attention mechanisms, outperform traditional models in capturing the nuanced sentiment expressed in tweets. Feature extraction techniques such as Bag-of-Words (BoW), TF-IDF are to represent the textual data in format suitable for machine learning algorithms. Investigated the effectiveness of different machine learning models including Logistic Regression, Bernoulli Naive Bayes, Support Vector Machine, K Nearest Neighbors, and Random Forest for sentiment classification. Evaluated using recall and F1 Score, experimental results show efficacy of proposed model for twitter data.