WITHDRAWN: A systematic study of sentiment analysis for social media data

Abstract

Abstract Sentiment Analysis is considered as the technique of defining and extracting human feelings through the unstructured text and is done through Natural Language Processing and Machine Learning. Machine Learning is a great way that helps in learning and training the datasets obtained from social media. It is the most ubiquitous strategy used in sentiment analysis. With machine learning, the lexicon-based, as well as the rule-based methods, can also be utilized. Analysis of sentiment is the best instrument to determine whether the evaluation is positive or negative. The various complications in the assessment of thoughts are one that the public does not always articulate feelings, in the same manner, which means a few expresses in the mode of providing scores and remarks and others in the form of phrases that do not convey any proper mindsets. The reviews are classified as positive, negative, and neutral. This review provides a synthesis of various prevailing sentiment analysis techniques employed on social media data. Different studies conducted in this area of research will be explored while discussing the algorithms developed. These include Naive Bayes, K-nearest method, Random Forest, Support Vector Machine, and Deep Learning. All these classification techniques were analyzed, with their issues and challenges.