Article Per Year (5 Year)
p-Index From 2019 - 2024
0.61
P-Index
This Author published in this journals
All Journal JOIV : International Journal on Informatics Visualization IJISTECH (International Journal Of Information System & Technology) IJISTECH
Andi Nur Rachman
Universitas Siliwangi Tasikmalaya, Indonesia
Computer Science & IT Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Engineering Social Sciences

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : IJISTECH (International Journal Of Information System

 1 
Comparison of the Multinomial Naive Bayes Algorithm and Decision Tree with the Application of AdaBoost in Sentiment Analysis Reviews PeduliLindungi Application Cecep Muhamad Sidik Ramdani; Andi Nur Rachman; Rizki Setiawan
IJISTECH (International Journal of Information System and Technology) Vol 6, No 4 (2022): Decembar
Publisher : Sekolah Tinggi Ilmu Komputer (STIKOM) Tunas Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/ijistech.v6i4.257

Abstract

One solution that the Indonesian government has implemented in controlling and tracking COVID-19 cases is using the PeduliLindungi application. User review data on the PeduliLindungi application is available on the Google Play Store, the data can be analyzed to determine the trend of public sentiment towards the PeduliLindungi application using sentiment analysis techniques. One of the methods used for sentiment analysis is machine learning, but in the machine learning method there is a problem, namely the relatively low level of accuracy. In this study, there are 2 machine learning algorithms that are used and compared, namely the Multinomial Naïve Bayes (MNB) and Decision Tree (DT) algorithms combined with the AdaBoost (AB) method to improve the accuracy of the PeduliLindungi application review data classification accuracy. In the experiment conducted, the tendency of public sentiment towards the PeduliLindungi application was 67% positive and 33% negative from a total of 8305 data. Multinomial Naïve Bayes before being combined with AdaBoost produces an average accuracy value of 83,7%, while Decision Tree produces an average accuracy value of 82,8%. After being combined, MNB+AB produces an average accuracy value of 88,8%, while the DT+AB method produces an average accuracy value of 84,1%. The use of AdaBoost can improve the accuracy of the Multinomial Naive Bayes algorithm and Decision Tree for the PeduliLindungi application review data classification process.
Co-Authors Cecep Muhamad Sidik Ramdani Euis Nur Fitriani Dewi Husni Mubarok Rama Edwinda Putra Rizki Setiawan