Sentiment Analysis using SVM Machine Learning Techniques on Social Media

Sentiment analysis is a branch of Natural Language Processing (NLP) and machine learning in which text is classified into two categories: positive and negative. Because the use of the internet and social media is growing at a rapid pace, the products created by these two are receiving far more client input than in the past. Text generated by social media, blogs, posts, and product reviews, among other places, has become the best-suited examples for consumer sentiment, delivering the best-suited notion for that particular product. As a result, the hybrid feature selection proposed in this paper is a combination of particle swarm optimization (PSO) and cuckoo search. The hybrid feature selection technique surpasses the standard technique due to the subjective nature of social media reviews. Support Vector Machine (SVM) classifier was used to examine performance criteria such as f-measure, recall, precision, and accuracy on the Twitter dataset, and it was compared to a convolution neural network. The proposed work outperforms the existing work, according to the findings of this paper's experiments based on various factors.