Klasifikasi Respons Terhadap Vaksinasi Covid-19 Berdasarkan Tweets Menggunakan Attention-Based Long Short Term Memory

Salsabila, Diva Zannuba (2022) Klasifikasi Respons Terhadap Vaksinasi Covid-19 Berdasarkan Tweets Menggunakan Attention-Based Long Short Term Memory. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06111740000060-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2024. Download (12MB) | Request a copy

Abstract

Media sosial memudahkan masyarakat dalam mendapatkan informasi dan menuangkan pendapat, saran atau kritiknya dalam peristiwa tertentu. Vaksinasi virus COVID-19 di Indonesia yang sedang hangat diperbicangkan dan mendapatkan beragam respons dari masyarakat baik pro maupun kontra, dapat dimanfaatkan untuk melakukan analisis terhadap respons tersebut. Untuk mendukung analisis tersebut, Tugas Akhir ini bertujuan untuk mengklasifikasikan respons dari masyarakat Indonesia terhadap vaksinasi COVID-19 menjadi tiga kelas yaitu negatif, netral, dan positif. Untuk proses klasifikasi respons tersebut, Tugas Akhir ini mengimplementasikan metode Attentional-based Long Short Term Memory atau A-LSTM. Disisi lain, Tugas Akhir ini juga mengimplementasikan Bidirectional Encoder Representation Transformer (BERT) sebagai metode pada proses tokenisasi untuk memperoleh representasi fitur dari data Tweet sehingga membantu proses pelatihan A-LSTM. Proses evaluasi dilakukan dengan menggunakan dataset Tweets Bahasa Indonesia dari media sosial Twitter dimulai dari diangkatnya isu vaksinasi COVID-19 di Indonesia. Hasil dari metode ini menunjukkan kinerja yang baik dengan nilai akurasi sebesar 82%.
================================================================================================
Social media helps people to get information and give opinion on certain events. Vaccination for COVID-19 in Indonesia which much talked about and gets various responses from public, both pros and cons, can be used to analyze the responses. To support this, this Final Project aims to classify responses of Indonesian people to COVID-19 vaccination into three classes: negative, neutral, and positive using Attention-based Long Short Term Memory (A-LSTM) algorithm. This Final Project also using Bidirectional Encoder Representation Transformet (BERT) as method to tokenize for getting feature representation of Tweet data to assist the training process of A-LSTM. The evaluation process was carried out using the Indonesian Tweets dataset from Twitter social media, starting with the issue of COVID-19 vaccination in Indonesia. The results of this method show good performance with an accuracy value of 82%.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Media Sosial, Twitter, Analisis Sentimen, Attention-based Long Short Term Memory (A-LSTM), Social Media, Twitter, Sentiment Analysis
Subjects:	H Social Sciences > HA Statistics > HA31.38 Data envelopment analysis. H Social Sciences > HD Industries. Land use. Labor > HD108 Classification (Theory. Method. Relation to other subjects ) Q Science > QA Mathematics > QA278.55 Cluster analysis Q Science > QA Mathematics > QA279 Response surfaces (Statistics). Analysis of covariance. Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis. Correspondence analysis (Statistics)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Diva Zannuba Salsabila
Date Deposited:	10 Feb 2022 05:20
Last Modified:	02 Nov 2022 02:01
URI:	http://repository.its.ac.id/id/eprint/93053

Actions (login required)

View Item