Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal ELINVO (Electronics, Informatics, and Vocational Education)
Kasyfurrahman, M. Akbar
Unknown Affiliation
Computer Science & IT Education Electrical & Electronics Engineering

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

Found 1 Documents
Search

 1 
Sentiment Analysis of Cooking Oil Prices in Indonesia Using the Long Short-Term Memory Method Mulyani, Evi Dewi Sri; Hidayat, Cepi Rahmat; Mufizar, Teuku; Sundari, Shinta Siti; Anwar, Dede Syahrul; Ruuhwan; Ma'ruf, Jamal; Kasyfurrahman, M. Akbar
Elinvo (Electronics, Informatics, and Vocational Education) Vol. 10 No. 1 (2025): May 2025 (In Press)
Publisher : Department of Electronic and Informatic Engineering Education, Faculty of Engineering, UNY

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21831/elinvo.v10i1.73972

Abstract

The policy that was intensively discussed from early January to March 2022 was related to the setting and lifting of the Highest Retail Price for cooking oil, which was proven by the busy news on television, print media and social media. Many responses or speculations arise as a result of this policy. The importance of research on public speculation or sentiment analysis is to create a system model and find out how the public responds to government policy after the Highest Retail Price for cooking oil is determined and revoked, which ranges from February 4 to March 31 2022, as a benchmark and material. government considerations in making policy. The data that was collected in the period after the Maximum Retail Price (MRP) was set amounted to 904 datasets, and after the repeal of the MRP, it amounted to 874. Research can function as a basis for completing important information to support public policy decisions. The data is trained to obtain an optimal model and can predict sentiment with the Long Short-Term Memory (LSTM) model. To get the best model, random parameter testing was carried out using 80% of the training data and 20% of the validation dataset. The test results in fairly good accuracy with the softmax activation function, with an accuracy of 82.34%.
Co-Authors Cepi Rahmat Hidayat Dede Syahrul Anwar, Dede Syahrul Evi Dewi Sri Mulyani Ma'ruf, Jamal Ruuhwan Shinta Siti Sundari Teuku Mufizar