EVALUATING MACHINE LEARNING CLASSIFIERS FOR SENTIMENT ANALYSIS ON SOCIAL MEDIA DATA

Abstract

Social media platforms have become integral parts of our daily lives, creating an overwhelming volume of user-generated content that reflects human emotions, thoughts, and experiences. Sentiment analysis, a powerful application of natural language processing and machine learning, has emerged as a revolutionary technology to analyze and interpret the emotional tone of this vast social media data. This noise was eliminated during preliminary processing of the data using noise elimination software. Researchers used a wide variety of methods in their studies. Multi-layer Perceptron (MLP) and Convolutional Neural Networks (CNN) were used to sort the data into categories. Based on the sentiment classification from Twitter data and consumer affairs website, these two classifications were compared to the other classed such as Support Vector Machine (SVM), Random Forest, Decision tree, Nave Bayes, etc. Results from the proposed study show that, compared to other Machine Learning Classifiers, Multi-layer Perceptron and Convolutional Neural Networks perform the best.

Keywords: Classifiers, Machine Learning, Twitter, Sentiment, Social media