ANALISIS SENTIMEN PADA DATA ULASAN APLIKASI PLN MOBILE MENGGUNAKAN VADER LEXICON DAN NAÃVE BAYES
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
FAJRI, MUHAMAD
Yosrita, Efy
Asri, Yessy
Yosrita, Efy
Asri, Yessy
Subject
Teknik Informatika
Datestamp
2023-06-09 07:29:01
Abstract :
PLN Mobile application is a digital application created by PT PLN (Persero) with the aim of providing electricity services through a mobile application. Reviews on Google Playstore have a rating of 1 to 5, but often users give ratings that are not in accordance with their reviews so that this is not enough to describe the quality of the application. The number of reviews or data reviews on the PLN Mobile application is so large that it will take time and time to read in its entirety. To find out the public opinion, a classification system is applied. Sentiment Analysis uses 1000 samples of review data from 67949 population data taken from January to June 2022 on the PLN Mobile Application. In this study, the process of collecting review data (Web Scraping), Text Preprocessing, data labeling, text classification and evaluation models was carried out. For the text classification method using a Lexicon-based approach, which will use a dictionary-based approach (Vader Lexicon) resulting in 469 positive sentiments, 447 negative sentiments, 84 neutral. From the results of the comparison of positive, neutral, and negative classes against 1000 samples of Vader Lexicon data with reviews based on the user's discrepancy rating, the rating is for the positive class of 20%, the neutral class of 39%, the negative class of 19%. Furthermore, the classification process is carried out using the nave Bayes method, for the distribution of test and training data the author uses a ratio of 90:10 for the split data. Evaluation model process using confusion matrix get accuracy 54 %
PLN Mobile application is a digital application created by PT PLN (Persero) with the aim of providing electricity services through a mobile application. Reviews on Google Playstore have a rating of 1 to 5, but often users give ratings that are not in accordance with their reviews so that this is not enough to describe the quality of the application. The number of reviews or data reviews on the PLN Mobile application is so large that it will take time and time to read in its entirety. To find out the public opinion, a classification system is applied. Sentiment Analysis uses 1000 samples of review data from 67949 population data taken from January to June 2022 on the PLN Mobile Application. In this study, the process of collecting review data (Web Scraping), Text Preprocessing, data labeling, text classification and evaluation models was carried out. For the text classification method using a Lexicon-based approach, which will use a dictionary-based approach (Vader Lexicon) resulting in 469 positive sentiments, 447 negative sentiments, 84 neutral. From the results of the comparison of positive, neutral, and negative classes against 1000 samples of Vader Lexicon data with reviews based on the user's discrepancy rating, the rating is for the positive class of 20%, the neutral class of 39%, the negative class of 19%. Furthermore, the classification process is carried out using the nave Bayes method, for the distribution of test and training data the author uses a ratio of 90:10 for the split data. Evaluation model process using confusion matrix get accuracy 54 %
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
