Analisa Kinerja RNN Menggunakan FastText Embedding terhadap Ulasan PeduliLindungi di Masa COVID-19

SHALEHAH, ADINDA SITI (2022) Analisa Kinerja RNN Menggunakan FastText Embedding terhadap Ulasan PeduliLindungi di Masa COVID-19. S1 thesis, Universitas Mercu Buana Jakarta.

[img]
Preview
 Text (HAL COVER)
01 COVER.pdf
Download (735kB) | Preview
[img]
Preview
 Text (ABSTRAK)
02 Abstrak.pdf
Download (53kB) | Preview
[img] Text (BAB I)
03 BAB 1.pdf
Restricted to Registered users only
Download (149kB)
[img] Text (BAB II)
04 BAB 2.pdf
Restricted to Registered users only
Download (196kB)
[img] Text (BAB III)
05 BAB 3.pdf
Restricted to Registered users only
Download (201kB)
[img] Text (BAB IV)
06 BAB 4.pdf
Restricted to Registered users only
Download (267kB)
[img] Text (BAB V)
07 BAB 5.pdf
Restricted to Registered users only
Download (277kB)
[img] Text (BAB VI)
08 BAB 6.pdf
Restricted to Registered users only
Download (452kB)
[img] Text (DAFTAR PUSTAKA)
09 Daftar Pustaka.pdf
Restricted to Registered users only
Download (109kB)
[img] Text (LAMPIRAN)
10 Lampiran.pdf
Restricted to Registered users only
Download (181kB)

Abstract

Abstract − PeduliLindung is an application designed by the Ministry of Communication and Information in collaboration with the Ministry of Health and the Ministry of State-Owned Enterprises with the aim of monitoring the spread of COVID-19. In addition, it can be used to register and view vaccine status. The application can be downloaded through the market app, one of which is the Google Play Store. PeduliLindung users are increasing day by day so there are pros and cons regarding the application system. Users submit their comments through the feedback feature on the Google Play Store in the form of reviews and ratings. The reviews given by users are quite varied, so a classification technique is needed to group reviews based on their class. Reviews are given two class labels, namely Positive and Negative. Reccurent Neural Network (RNN) is a proposed algorithm to handle text classification because its most famous architecture is used for text processing. This study examines the performance of RNN with three variants, namely SimpleRNN, Long Short Term Memory, and Bidirectional-LSTM with the addition of FastText embedding as input to improve algorithm performance. The results showed that text classification using RNN with the addition of FastText obtained good and quite accurate results. The highest accuracy obtained by the LSTM and Bi-LSTM models with the addition of FastText is 89% and SimpleRNN 88%. Key words: COVID-19, Classification, RNN, LSTM, Bi-LSTM, FastText. Abstrak − PeduliLindungi merupakan aplikasi rancangan Kementerian Komunikasi dan Informatika bekerjasama dengan Kementerian Kesehatan dan Kementerian Badan Usaha Milik Negara dengan tujuan untuk memantau penyebaran COVID-19. Selain itu, dapat digunakan untuk mendaftar dan melihat status vaksin. Aplikasi tersebut dapat diunduh melalui market app salah satunya Google Play Store. Pengguna PeduliLindungi kian hari kian meningkat sehingga adanya pro dan kontra mengenai sistem aplikasi. Pengguna menyampaikan komentarnya melalui fitur feedback yang ada di Google Play Store berupa ulasan dan rating. Ulasan yang diberikan pengguna cukup bervariasi maka dibutuhkan teknik klasifikasi untuk mengelompokan ulasan berdasarkan kelasnya. Ulasan diberi dua label kelas yaitu Positif dan Negatif. Reccurent Neural Network (RNN) merupakan algoritma yang diusulkan untuk menangani klasifikasi teks karena arsitekturnya yang paling terkenal digunakan untuk pemrosesan teks. Penelitian ini menguji kinerja RNN dengan tiga variannya yaitu SimpleRNN, Long Short Term Memory, dan Bidirectional-LSTM dengan penambahan FastText embedding sebagai input untuk meningkatkan kinerja algoritma. Hasil penelitian menunjukan bahwa klasifikasi teks menggunakan RNN dengan penambahan FastText memperoleh hasil yang baik hasil yang baik dan cukup akurat. Akurasi tertinggi diperoleh oleh model LSTM dan Bi-LSTM dengan penambahan FastText adalah 89% dan SimpleRNN 88%. Kata Kunci : COVID-19, Klasifikasi, RNN, LSTM, Bi-LSTM, FastText.

Item Type: Thesis (S1)
Call Number CD: FIK/INFO.22 127
Call Number: SIK/15/22/040
NIM/NIDN Creators: 41518010031
Uncontrolled Keywords: COVID-19, Klasifikasi, RNN, LSTM, Bi-LSTM, FastText.
Subjects: 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum
000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 001 Knowledge/Ilmu Pengetahuan
000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 004 Data Processing, Computer Science/Pemrosesan Data, Ilmu Komputer, Teknik Informatika
Divisions: Fakultas Ilmu Komputer > Informatika
Depositing User: WADINDA ROSADI
Date Deposited: 19 Oct 2022 05:46
Last Modified: 19 Oct 2022 05:46
URI: http://repository.mercubuana.ac.id/id/eprint/70595

Actions (login required)

View Item View Item
UMB Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.