Sentiment Analysis Using Machine Learning A Comparative Study

• Published in: Advances in distributed computing and artificial intelligence journal Vol. 12; p. e26785
• Main Authors: Singh, Neha, Jaiswal, Umesh Chandra
• Format: Journal Article
• Language: English
• Published: Salamanca Ediciones Universidad de Salamanca 01.01.2023
• Subjects: Algorithms; Comparative studies; Data mining; Digital media; Machine learning; Performance evaluation; Regression models; Sentiment analysis; Social networks
• ISSN: 2255-2863; 2255-2863
• DOI: 10.14201/adcaij.26785

In recent years, sentiment analysis on social media, including Facebook, Twitter and blogs, has grown in popularity. Social media generate large amounts of information, and this has contributed to the growth of sentiment analysis as a field of research. This study demonstrates that sentiment analysis has been thoroughly researched in previous years, and numerous methods have been designed and evaluated. Nevertheless, there is still much room for improvement. This paper reviews the state of art in sentiment analysis. Various machine learning procedures for sentiment analysis are discussed, their potential to increase the level of the analysis accuracy is underscored. This paper introduces sentiment analysis types, methodologies, applications, challenges, and a comparative study of machine learning and sentiment analysis approaches. Performance evaluation parameters, for sentiment analysis, have also been tested and compared using different machine learning classifiers. Performance evaluation points to logistic regression as the model that achieves the best result. In the future, a method that is easy, versatile, and practicable, should be offered as opposed to existing machine learning methods, and more work should be put into improving the algorithms’ performance.