<![CDATA[Indihome is a digital service such as the internet that can be used at home, landlines and interactive TV. However, because it is so extensive, Indihome has received a lot of criticism because the internet connection is rarely stable. Therefore, a sentiment analysis in the field of was carried out data mining on customers Indihomeon Twitter social media which consisted of 1350 data and filtered into 1309 data because a lot of data indicated duplicates. In this study, researchers used the methods Random Forest and Gradient Boosted Trees (GBT). This research was conducted using tools Rapidminer version 9.8. Research shows that sentiment analysis on Indihome services using the method Random Forest achieves an accuracy of 99.54% with class precision for pred. negative is 99.92%, pred positive is 25.00%, and pred. neutral is 60.00%. Then the GBT method has an accuracy rate of 99.31% with a precision class ofn for pred. negative is 99.46%, pred. positive is 0.00%, and pred. neutral is 0.00%. So it can be concluded that the Random Forest method is a better method when compared to GBT.]]>
Copyrights © 2021
