1. Sentiment analysis of multi-social media platform about medical services using support machine vector.
- Author
-
Nainggolan, Cindy H., Amalia, Amalia, Sharif, Amer, and Lydia, Maya Silvi
- Subjects
- *
SENTIMENT analysis , *SUPPORT vector machines , *SOCIAL media , *USER-generated content , *LOCAL mass media , *INDONESIAN language , *CLASSIFICATION algorithms - Abstract
Social media contains robust source information of people's viewpoints because everyone can give their opinion independently. Due to various types of social media, there are different methods for scrapping data because different social media have different regulations. This study aims to build sentiment analysis with data sources in Bahasa Indonesia (Indonesian Language) from various social media platforms such as Twitter, Facebook, and Tiktok. A case study from this research regarding health workers' services for the community on social media with the Support Vector Machine (SVM) classification algorithm. This study used the hashtag #Tenagamedis (in English: health workers). The methodology of this study contains data collecting, filtering, pre-processing, data labeling, and classification. The crawling data get a total of 4223 comments with a composition of 617 from Twitter, 1650 from Facebook, and 2001 from Tiktok. The data from Tiktok get the best accuracy which is 91%. Based on the test results, the amount of data and the balance of the training data determine the classification accuracy. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF