Article Per Year (5 Year)
p-Index From 2019 - 2024
0.23
P-Index
This Author published in this journals
All Journal JURNAL MEDIA INFORMATIKA BUDIDARMA
Ganjar Tata Pangestu
Universitas Ahmad Dahlan, Yogyakarta
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Sentiment Analysis Tweet Pilkada 2020 Saat Pandemik COVID-19 di Media Sosial Twitter Menggunakan Metode 1D Convolutional Neural Network Ganjar Tata Pangestu; Miftahurrahma Rosyda
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 6, No 2 (2022): April 2022
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v6i2.3765

Abstract

Pilkada 2020 is a debate since it takes place in the midst of the COVID-19 pandemic. The emergence of comments from several social media such as Twitter. There are various public opinions that agree that the Pilkada will still be held, there are also other public opinions that support the postponement of the Pilkada until the COVID-19 pandemic ends. These different opinions require Sentiment Analysis which aims to obtain or find out the general opinion of the 2020 Regional Head Election during the Coronavirus pandemic. A total of 200 data in the form of tweets which are divided into 2 data, namely 20% test data and 80% training data obtained by retrieving data from Twitter using the twint library, based on predetermined keywords. The resulting data set is classified into three classes namely positive, neutral, and negative. In this test or research, deep learning uses the Convolutional Neural Network classification, because it has been proven effective in the case of natural language processing and can get good results in grouping sentences. From this research, the accuracy result is 72.50% with the epoch used is 25 epoch. From the increase in epochs, there is an increase in accuracy of 7-12% from the previous epoch variations.
Co-Authors Miftahurrahma Rosyda