Machine Learning Approaches for Sentiment Analysis in Social Media Data

As a crucial component of Natural Language Processing, sentiment analysis autonomously identifies and categorizes the positive and negative emotions reflected in social media text data. Here, the proposed work examines the viability of using Support Vector Machines (SVM) as a technique for machine learning to decipher online comments. Analysis of public opinion is crucial in the information age. Due to its nonlinear separability and efficacy with high-dimensional data, SVM is well-suited for this task. The bag-of-words technique is utilized for data acquisition, cleaning, and feature extraction. The SVM model is trained using cross-validation and hyperparameter adjustment, resulting in a 97% accuracy rate. Compared to other algorithms, the proposed SVM technique is preferable to the current method. The proposed work sheds light on how to assess the attitudes of social media users, whose findings may have significant implications for industries, including advertising, politics, and brand public opinion.