Twitter Sentiment Analysis For Feature Extraction Using Support Vector Machine (SVM) With TF-IDF

Abstract

The goal of this research is to perform sentiment analysis on Twitter data by extracting relevant features using the Hybrid Support Vector Machine (SVM) with TF-IDF (term frequency-inverse document frequency) weighting scheme. Sentiment analysis involves analyzing the emotional tone of a piece of text, and in this case, is specifically interested in classifying tweets as positive, negative, or neutral. To achieve this, using a yahoo stock market dataset of tweets using the Twitter API and preprocess the data by removing stop words, special characters, and URLs. Proposed the SVM-TF-IDF to extract features from the preprocessed tweets calculates the importance of each word in a tweet by taking into account both its frequency within the tweet and its rarity across the entire stock dataset. The research aims to demonstrate the effectiveness of the SVM-TF-IDF for feature extraction in sentiment analysis tasks and to provide insights into the sentiment trends and patterns present in Twitter data.