<![CDATA[Technological developments, especially in the internet and social media, could be a very important research subject in obtaining information, because of the large amount of information in a text found on social media. In recent years, there has been an increase in research about sentiment analysis on text reviews and tweets in order to determine the polarity generated by social media. There are still few studies that apply the deep learning method with the Long Short-Term Memory (LSTM) algorithm to analyze multiclass sentiments in Indonesian-language texts. This study aims to analyze positive and negative emotions in social media texts using the information classification approach in the text and dividing them into 8 different classes using the LSTM method. The dataset is directly taken and collected from users' posts on social media. In testing the LSTM method, the calculation of the accuracy, exactness, review, f-measure values is generated. The results of the processing of the LSTM method show quite well with 5 trials with the highest accuracy value of 91.9% and the average value of multiclass getting 89.45% results.]]>
Copyrights © 2021
