IMPLEMENTASI ALGORITMA RECURRENT NEURAL NETWORK DENGAN ARSITEKTUR LONG SHORT-TERM MEMORY DALAM KLASIFIKASI SENTIMEN PENYELENGGARAAN FIFA WORLD CUP 2022 QATAR

HERDIANSYAH, MUHAMMAD WAHYU and Utami, Alvi Syahrini and Darmawahyuni, Annisa (2023) IMPLEMENTASI ALGORITMA RECURRENT NEURAL NETWORK DENGAN ARSITEKTUR LONG SHORT-TERM MEMORY DALAM KLASIFIKASI SENTIMEN PENYELENGGARAAN FIFA WORLD CUP 2022 QATAR. Undergraduate thesis, Sriwijaya University.

	Text RAMA_55201_09021281924038.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021281924038_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (7MB) | Request a copy
	Text RAMA_55201_09021281924038_0022127804_8968340022_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (988kB)
	Text RAMA_55201_09021281924038_0022127804_8968340022_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (351kB) | Request a copy
	Text RAMA_55201_09021281924038_0022127804_8968340022_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (171kB) | Request a copy
	Text RAMA_55201_09021281924038_0022127804_8968340022_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (398kB) | Request a copy
	Text RAMA_55201_09021281924038_0022127804_8968340022_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (450kB) | Request a copy
	Text RAMA_55201_09021281924038_0022127804_8968340022_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (13kB) | Request a copy
	Text RAMA_55201_09021281924038_0022127804_8968340022_07_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (81kB) | Request a copy
	Text RAMA_55201_09021281924038_0022127804_8968340022_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (7kB) | Request a copy

Abstract

The text classification of social media posts poses certain challenges due to the linguistic diversity that exists, making it an interesting problem to research. This research was conducted to classify sentiments towards the opinion of the community. The use of the RNN algorithm which stores previous data for the latest output results is suitable in this study because it can process sequential data in the form of text well in accordance with this study, although it is unable to store these data for a long time. The LSTM architecture can complement the shortcomings of RNNs which cannot predict a word based on past information that has been stored for a long time. Thus, LSTM can make the data classification process more efficient because it is able to remember information that has been stored for a long time and delete information that is no longer relevant. The data used in this study amounted to 440 data with 200 positive classes, 150 neutral classes, and 90 negative classes. The process of making the model was carried out in several scenarios by tuning the learning rate and batch size parameters, each of which had a value of 0.1 to 0.0001 and 8 to 64. The best model that was obtained was a model with a learning rate value = 0.001 and a batch size = 16 which has a high level of accuracy. by 69.32%.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Klasifikasi Sentimen, RNN, LSTM
Subjects:	Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Muhammad Wahyu Herdiansyah
Date Deposited:	28 Jul 2023 07:51
Last Modified:	28 Jul 2023 07:51
URI:	http://repository.unsri.ac.id/id/eprint/123339

Actions (login required)

View Item