Analisis Sentimen Tentang Berita Terkait Risiko Industri di Indonesia Pada Masa Pandemi COVID-19 Menggunakan Metode Hybrid GRU-SVM

Chofiyya, Violeta Nur (2021) Analisis Sentimen Tentang Berita Terkait Risiko Industri di Indonesia Pada Masa Pandemi COVID-19 Menggunakan Metode Hybrid GRU-SVM. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

	Text organized.pdf Restricted to Repository staff only Download (2MB) | Request a copy
	Text organized.pdf Restricted to Repository staff only Download (2MB) | Request a copy
	Text organized.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy
	Text 06111640000095-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Dengan diberlakukan kebijakan Pembatasan Sosial Berskala Besar (PSBB) oleh Pemerintah Indonesia, banyak sektor mengalami keterpurukan seperti halnya pada sektor Industri. Salah satu upaya untuk membangkitkan kondisi terpuruk sektor industri adalah dengan mengetahui beberapa risiko industri selama masa pandemi COVID19. Dengan memanfaatkan media sosial, Tugas Akhir ini bertujuan untuk menganalisis sentimen pendapat masyarakat terkait risiko industri selama masa pandemi COVID19. Analisis sentimen pada penelitian ini dilakukan dengan menggunakan metode hybrid antara Gated Recurrent Unit (GRU) dan Support Vector Machine (SVM) yang disingkat menjadi Hybrid GRUSVM. Secara singkat, metode GRU digunakan untuk membangkitkan fitur semantik teks data dari media sosial. Selanjutnya, SVM mengklasifikasikan fitur semantik tersebut menjadi positif, netral atau negatif. Metode Hybrid GRUSVM diimplementasikan pada data Twitter dan berita online terkait risiko industri selama pandemi COVID19 di Indonesia, dan dibandingkan dengan metode dasar yaitu GRU dan SVM. Hasil eksperimen diperoleh bahwa SVM mampu mengalahkan baik metode GRU dan Hybrid GRUSVM.
====================================================================================================
As a largescale social restriction enacted by government, many sectors was have experienced a slump as in the industrial sector. One of the efforts to revive the slumping condition of the industrial sector is to know some industrial risks during the COVID19 pandemic. By utilizing social media, this Final Project aims to analyze public opinion sentiment regarding industrial risks during the COVID19 pandemic. Sentiment analysis in this study was conducted using a hybrid method between the Gated Recurrent Unit (GRU) and the Support Vector Machine (SVM) which is abbreviated as Hybrid GRUSVM. Briefly, the GRU method is used to generate semantic features of data text from social media. Furthermore, SVM classifies these semantic features into positive, neutral or negative. The Hybrid GRUSVM method is implemented on Twitter data and online news related to industrial risks during the COVID19 pandemic in Indonesia, and compared to the basic methods, namely GRU and SVM. The experimental results show that SVM is able to beat both the GRU and Hybrid GRUSVM methods.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Risiko Industri, Analisis Sentimen, GRUSVM, Industry Risk, Sentiment Analysis
Subjects:	R Medicine > RA Public aspects of medicine > RA644.C67 COVID-19 (Disease) T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL221.5 Hybrid Vehicles. Hybrid cars
Divisions:	Faculty of Mathematics and Science > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Violeta Nur Chofiyya
Date Deposited:	03 Sep 2021 03:26
Last Modified:	03 Sep 2021 03:26
URI:	http://repository.its.ac.id/id/eprint/91260

Actions (login required)

View Item