Sentiment Analysis Siap Kerja Training Services with Naive Models Bayes and SVM

Abstract

Training service Siap Kerja has an important function in improving the quality of human resources and developing a competent workforce in various industrial sectors. To ensure trainee success and satisfaction, sentiment analysis in trainee reviews is essential to evaluate the strengths and weaknesses of the service offered. This research intends to compare the performance of algorithms Support Vector Machine (SVM) and Naive Bayes in sentiment analysis of trainee reviews on pages service training Siap Kerja. Review data was collected using web scraping on the Siap Kerja training service page by collecting ratings and reviews of training participants after completing the training they attended on the page training service Siap Kerja. Methods of data pre-processing, data sharing, feature extraction, model training, model evaluation, and results analysis were used in this research. The research results show that the SVM model has higher accuracy (0.93) compared to the Naive model Bayes (0.80). Additionally, positive sentiment was found at 99%, negative at 0.3%, and neutral at 0.7%. ased on the results of this research, it is recommended that the SVM model be used in sentiment analysis in the context of Siap Kerja training services and that the model be optimized to improve performance in classifying reviews with negative or neutral feelings.