Sentiment Analysis for Real-Time Micro Blogs using Twitter Data
Reshma Banu, G. F. Ali Ahammed, G. Divya, V. Dinesh Reddy, Nuthanakanti Bhaskar, Murali Kanthi
Abstract
The basic purpose of sentiment analysis is to determine how someone feels when they comment or express their feelings or emotions. Positive, neutral, and negative emotions are the three categories into which emotions are divided. Everyone will use and apply this analysis on social media; online; everyone expresses their opinions by clicking on the like, remark, or share buttons. Using the Random Forest, SVM, and Nave Bayes algorithms, the Twitter tweets in this study were identified as positive or negative, with F1-Scores of 0.224, 0.410, and 0.702, respectively, and accuracy values of 50%, 52%, and 73%.
Topics & Concepts
Sentiment analysisFeelingSocial mediaNaive Bayes classifierComputer scienceSupport vector machineRandom forestPsychologyArtificial intelligenceWorld Wide WebSocial psychologySentiment Analysis and Opinion MiningAdvanced Text Analysis TechniquesWeb Data Mining and Analysis