Claim Missing Document
Check
Articles

Found 1 Documents
Search

Analisis Sentimen Complain dan Bukan Complain pada Twitter Telkomsel dengan SMOTE dan Naïve Bayes Budi Kurniawan; Achmad Suwarisman; Iis Afriyanti; Aditya Wahyudi; Dedi Dwi Saputra
Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) Vol 7 No 1 (2023): JANUARY-MARCH 2023
Publisher : KITA Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jtik.v7i1.691

Abstract

This analysis aims to find out the public sentiment towards Telkomsel posted on Indonesian twitter, which makes market research on public opinion very useful. The dataset was taken from Twitter social media in a query Indonesian by crawling method using the RapidMiner application and the result of crawling the data set there were 1000 tweets with sentiment complaints and not complaints. Therefore, from 1000 tweets, preprocessing will be carried out with the SMOTE Upsampling and Naivebayes methods as well as several filtering such as transform case, tokenize, tokenize (by length) stemming filters and stopwords so that the data can stay in words and there is a balance in the sentiment on the dataset. It can be concluded that in the classification of sentiment there is a balance between complaints and non-complaints as many as 581. Where the accuracy rating level is 81.58%, the precision assessment is 86.82% and the recall assessment is 74.87 and the resulting AUC is 0.803.