Aam Amaningsih Jumhur, Irma Permata Sari, Shandy Aditya, Anter Venus
People's opinions regarding SMEs in Subang Regency are extensively expressed in the comment sections found on social media. This can be applied to observe the emotional tendencies of the public towards a particular topic of discussion, thus providing an insight into the genuine responses from the community. The research findings include the performance evaluation of a sentiment analysis classification model for SMEs Subang on YouTube comments using the Support Vector Machine and TF-IDF methods to assess their accuracy. After completing the Pre-Processing, TF-IDF word weighting, and SVM classification stages, the data was split with 90% used for training and 10% for testing. The model's performance was evaluated using both the Confusion Matrix and K-Fold Cross Validation methods. The evaluation results without using Cross Validation showed an accuracy of 81.81%, a Precision of 71.43%, a Recall of 100%, and an f1 score of 83.33%. However, after performing validation using K-Fold Cross Validation with 7-fold testing, the average accuracy achieved was 41.67 %, the average Precision was 42.14 % the average Recall was 42.14 %, and the average f1 score was 42.14 %. © 2023 IEEE.
Jakarta State Universiy, Departement of Mechanical Education, Jakarta, Indonesia; Jakarta State Universiy, Information System and Technology Department, Jakarta, Indonesia; Jakarta State Universiy, Department of Business Digital, Jakarta, Indonesia; Universitas Pembangunan Nasional Veteran Jakarta, Informatic Departement, Jakarta, Indonesia
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock