Sentiment analysis is synonymous with opinion mining, is a type of data mining that refers to the analysis of data obtained from microblogging sites, social media updates, online news reports, user reviews, etc., to study people's sentiments towards an event, organization, product, brand, people and more. In this work, sentiment classification is carried out into several classes. The proposed methodology based on KNN classification algorithm shows improvement over one of the existing methodologies based on SVM classification algorithm. The data used for analysis was taken from Twitter, which is the most popular microblogging site. Source data has been extracted from Twitter. N-Gram modeling technique has been used for feature extraction and k-nearest neighbor supervised machine learning algorithm has been used for sentiment classification. The performance of the proposed and existing techniques is compared in terms of accuracy, precision, and recall. It has been analyzed and concluded that the proposed technique performs better in terms of all standard evaluation parameters.
Copyrights © 2024