Penerapan Weighted Word Embedding pada Pengklasifikasian Teks Berbasis Recurrent Neural Network untuk Layanan Pengaduan Perusahaan Transportasi
Submission Date: 2020-08-20 16:57:40
Accepted Date: 2021-08-16 13:35:47
Abstract
Twitter menjadi salah satu media sosial yang paling sering dan paling populer digunakan oleh perusahaan sebagai penyedia layanan pelanggan perusahaan. Adanya ribuan cuitan yang dapat masuk dalam setiap hari, tentu akan merepotkan operator layanan untuk mengkategorikan jenis berbagai cuitan tersebut, lebih-lebih jika proses pemilahan kategori cuitan harus dilakukan secara manual. Dalam Tugas Akhir ini, kategoriasi cuitan secara otomatis dibangun dan diimplementasi¬kan menggunakan model klasifikasi berbasis recurrent neural network (RNN) yang dikombinasikan dengan model weighted word embedding (WWE). RNN merupakan salah satu jenis jaringan syaraf tiruan yang populer dan banyak digunakan dalam persoalan klasifikasi, sedangkan WWE merupakan metode yang memungkinkan untuk meng-hubungkan kata-kata yang serupa dengan mengukur jarak semantik antara vektor yang disematkan pada kata tersebut dan memberikan bobot yang berbeda pada setiap kata pada suatu kelas tertentu. Implementasi model penggabungan RNN dan WWE diuji coba menggunakan data pengaduan di perusahaan transportasi untuk data cuitan pada tahun 2015-2016. Hasil uji coba menunjukkan bahwa implementasi WWE baik yang menggunakan model FastText (Weighted FastText) maupun model Word2Vec (Weighted Word2Vec) memberikan hasil yang lebih baik dibandingkan dengan hasil kinerja yang menggabungkan RNN dan model word embedding biasa. Dengan menggunakan metode evaluasi berbasis 10-fold cross validation, model gabungan RNN-Weighted FastText dan RNN-Weighted Word2Vec berturut-turut memberikan hasil akurasi sebesar 88,2% dan 87,5%. Di lain pihak, dengan menggunakan metode evaluasi yang sama, model gabungan RNN-FastText dan RNN-Word2Vec memberikan hasil akurasi yang sama sebesar 83,4%.
Keywords
FastText; Klasifikasi Teks; Layanan Pengaduan Transportasi; Recurrent Neural Network; Twitter; Weighted Word Embedding; Word2Vec.
CC Licencing
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
Refbacks
- There are currently no refbacks.
Jurnal Sains dan Seni ITS by Lembaga Penelitian dan Pengabdian Kepada Masyarakat, LPPM-ITS is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Based on a work at https://ejurnal.its.ac.id/index.php/sains_seni.