SRI MULYANI, - IMPLEMENTATION OF THE NAIVE BAYES CLASSIFIER ALGORITHM FOR CLASSIFICATION OF COMMUNITY SENTIMENT ABOUT DEPRESSION ON YOUTUBE. IMPLEMENTATION OF THE NAIVE BAYES CLASSIFIER ALGORITHM FOR CLASSIFICATION OF COMMUNITY SENTIMENT ABOUT DEPRESSION ON YOUTUBE, 3 (6). ---. ISSN 2723-3863

Text (SRI MULYANI) SRI MULYANI.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Depression is a disease that knows no age, gender and social status. WHO states that more than 264 million people suffer from depression, people with depression will continue to grow if public knowledge about mental health is still low, especially in Indonesia. This can be known from the way the community responds to a case. This study aims to determine public sentiment towards people with depression by classifying comments using the Niave Bayes Classifier (NBC) algorithm and adding the Term Frequency-inverse Document Frequency (TF- IDF) method as a feature extraction method. Sentiment used as data is obtained from YouTube comments on several news media accounts such as tvOneNews, Kompas TV, Tribunnews, Official iNews, VIVACOID, CNN Indonesia and Tribun Jateng, so that 4783 data are obtained with training data of 3826 and 957 testing data. This sentiment was analyzed by giving three classes, namely positive, neutral and negative. The results of the sentiment analysis were dominated by positive sentiment of 93.31%, followed by negative comments of 6.68% while neutral sentiment was 0%, and the accuracy of the NBC Algorithm was 84.11%. Keywords: Depression, NBC, TF-IDF, YouTube

Actions (login required)

View Item