Sentiment Analysis for Transformer-based approach using Machine Learning Algorithms

Machine learning classifiers are now widely used across the world, transforming a variety of large and small scale industries. The number of algorithms—each allocated for specific tasks—can make it not simple for both beginners and experts to find the most appropriate model. Sentiment to be analyzed for human safety. This research tackles that challenge by experimenting with an in-depth analysis using Machine learning techniques and the efficient Transformers library. By conducting researchers' comments from existing studies, we proposed a novel method that maps classifiers performance based on descriptive adjectives. Machine learning algorithms producing effective highly accurate results in prediction. Accuracy is one of the performance metric to analyze performance of any machine learning classifiers Our proposed model shows sentence into positive, negative and neutral using Decision Tree(DT), Logistic regression, Naive bayes, Random forest (RF), Support Vector Machine(SVM) achieved 97% and K nearest neighbour(KNN).The proposed research is showing innovative in prediction of sentiment analysis compared to existing methods.

Hali tarjima qilinmagan