{"title":"Sentiment Analysis of Positive and Negative of YouTube Comments Using Naïve Bayes – Support Vector Machine (NBSVM) Classifier","authors":"Abbi Nizar Muhammad, S. Bukhori, P. Pandunata","doi":"10.1109/ICOMITEE.2019.8920923","DOIUrl":null,"url":null,"abstract":"Sentiment analysis on the YouTube video comments is a process of understanding, extracting, and processing textual data automatically to obtain sentiment information contained in one sentence of YouTube video comment. Text mining approach becomes the best alternative to interpret the meaning of each comment. The classification of positive and negative content becomes very important for the YouTube user to assess how meaningful the content that has been published is based on user opinion. Naïve Bayes and Support Vector Machine is extensively used as a basic line in tasks related to texts but the performance varies significantly in all variants, features, and numbers of data collection. Naïve Bayes is very good in classifying texts with the small number of data and document snippets while Support Vector is very good in classifying texts with relatively many numbers of data or full-length document. The combination of Naïve Bayes and Support Vector Machine produces better accuracy level and stronger performance with the use of a 7:3 scale of data that is 70% training data and 30% testing data. By producing the highest performance test values, namely precision of 91%, recall of 83% and flscore of 87%.","PeriodicalId":137739,"journal":{"name":"2019 International Conference on Computer Science, Information Technology, and Electrical Engineering (ICOMITEE)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"21","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 International Conference on Computer Science, Information Technology, and Electrical Engineering (ICOMITEE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICOMITEE.2019.8920923","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 21
Abstract
Sentiment analysis on the YouTube video comments is a process of understanding, extracting, and processing textual data automatically to obtain sentiment information contained in one sentence of YouTube video comment. Text mining approach becomes the best alternative to interpret the meaning of each comment. The classification of positive and negative content becomes very important for the YouTube user to assess how meaningful the content that has been published is based on user opinion. Naïve Bayes and Support Vector Machine is extensively used as a basic line in tasks related to texts but the performance varies significantly in all variants, features, and numbers of data collection. Naïve Bayes is very good in classifying texts with the small number of data and document snippets while Support Vector is very good in classifying texts with relatively many numbers of data or full-length document. The combination of Naïve Bayes and Support Vector Machine produces better accuracy level and stronger performance with the use of a 7:3 scale of data that is 70% training data and 30% testing data. By producing the highest performance test values, namely precision of 91%, recall of 83% and flscore of 87%.