A STUDY AND DEVELOPMENT OF APPLICATION ON SENTIMENT ANALYSIS

Abstract

A subfield of natural language processing called sentiment analysis is concerned with locating and obtaining subjective data from text. Analysing and categorising the emotional tone or polarity indicated in text—such as reviews, social media postings, news stories, and consumer feedback—is its primary goal. A subfield of natural language processing called sentiment analysis is concerned with locating and obtaining subjective data from text. Analysing and categorising the emotional tone or polarity indicated in text—such as reviews, social media postings, news stories, and consumer feedback—is its primary goal. Conversely, machine learning-based techniques employ algorithms to learn from a labelled dataset and categorise fresh data according to patterns seen in the training data. Hybrid methods combine both approaches to achieve better accuracy and coverage. Machine learning models, particularly Support Vector Machines (SVM), were utilized for sentiment analysis, involving the conversion of text data into numerical feature vectors and learning a hyperplane to classify sentiments accurately. Keywords: Sentiment Analysis, Machine Learning, HTML, Support Vector Machines, Bootstrap, Visual Studio Code.