ANALISIS SENTIMEN BERBASIS ASPEK MENGGUNAKAN METODE RANDOM OVER SAMPLING DAN RANDOM FOREST (Studi Kasus : Ulasan Pengunjung Objek Wisata Kabupaten Lamongan di Google Maps)

Ainandiyah, Anita (2023) ANALISIS SENTIMEN BERBASIS ASPEK MENGGUNAKAN METODE RANDOM OVER SAMPLING DAN RANDOM FOREST (Studi Kasus : Ulasan Pengunjung Objek Wisata Kabupaten Lamongan di Google Maps). Other thesis, UPN "Veteran" Yogyajarta.

	Text ABSTRAK.pdf Download (49kB)
	Text COVER.pdf Download (181kB)
	Text DAFTAR ISI.pdf Download (72kB)
	Text Laporan_TA_Anita Ainandiyah_123180029.pdf Restricted to Repository staff only Download (2MB)
	Text DAFTAR PUSTAKA.pdf Download (180kB)
	Text Halaman Pengesahan Pembimbing.pdf Download (219kB)

Abstract

vii
ABSTRAK
Fitur ulasan daring, seperti fitur Google Review di Google Maps, sangat penting bagi
industri pariwisata saat ini. ulasan dan umpan balik dari wisatawandapat dimanfaatkan
untuk meningkatkan kualitas pelayanan, memperbaiki pengalaman wisata, dan memperkuat
reputasi destinasi. Berbagai cara yang dapat dilakukan untuk penarikan ulasan dari
pengunjung di suatu fitur ulasan salah satunya adalah dengan menggunakan Analisis
Sentimen. Sudah terdapat beberapa penelitian pada ulasan objek wisata, namun namun tahap
klasifikasi masih pada level dokumen sehingga hasil masih kurang detail karena klasifikasi
dilakukan berdasarkan keseluruhan dokumen dan tidak mengidentifikasi aspek yang
dibicarakan secara spesifik, sehingga perlu dilakukan analisis sentimen berbasis aspek yang
dapat mengolah data ulasan untuk membedakan opini negatif dan opini positif berdasarkan
kategori aspek yang dibahas.
Pada penelitian ini dilakukan analisis sentimen berbasis aspek pada ulasan
Pengunjung objek wisata di Kabupaten Lamongan menggunakan metode Random Forest,
dengan 2 skenario. Skenario 1 menggunakan Random Forest tanpa Random Over Sampling,
skenario 2 menggunakan Random Forest dengan Random Over Sampling. Hasil pengujian
memberikan model terbaik dari skenario 2 dengan rata-rata skor Balanced Accuracy 89.13%,
F1-Score 88.37 % dan G-Mean 88.97 %. Berdasarkan hasil pengujian membuktikan bahwa
Random Over Sampling dapat meningkatkan kinerja model Random Forest.
Keywords : Analisis Sentimen Berbasis Aspek, Objek Wisata, Google Maps, Random Forest,
Random Over Sampling
viii
ABSTRACT
Online review features, such as the Google Review feature on Google Maps, are very
important for tourism industry today. Reviews and feedback from tourists can be utilized to
improve service quality, enhance tourist experience, and strengthen destination reputation.
Various ways can be done to attract reviews from visitors in a review feature, one of which
is by using Sentiment Analysis. There has been several reseacrh on tourist attraction reviews,
but the classification stage is still at the document level so the results are still less detailed
because the classification is based on the entire document and does not identify the aspects
discussed specifically, so it is necessary to do an aspect-based sentiment analysis that can
process review data to distinguish negative opinions and positive opinions based on the
category of aspects discussed.
This study aims to do aspect-based sentiment analysis on Visitor reviews of tourist
attractions in Lamongan Regency using the Random Forest method, with 2 scenarios.
Scenario 1 uses Random Forest without Random Over Sampling, scenario 2 uses Random
Forest with Random Over Sampling. The test results provide the best model from scenario
2 with an average value of Balanced Accuracy 89.13%, F1-Score 88.37% and G-Mean
88.97%. Based on the test results, it proves that Random Over Sampling can improve the
performance of the Random Forest model.
Keywords: Aspect-Based Sentiment Analysis, Tourism Objects, Google Maps, Random
Forest, Random Over Sampling

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Aspect-Based Sentiment Analysis, Tourism Objects, Google Maps, Random Forest, Random Over Sampling
Subjects:	Z Bibliography. Library Science. Information Resources > ZA Information resources
Divisions:	Faculty of Engineering, Science and Mathematics > School of Engineering Sciences
Depositing User:	Eko Yuli
Date Deposited:	08 Jun 2023 01:55
Last Modified:	08 Jun 2023 01:55
URI:	http://eprints.upnyk.ac.id/id/eprint/35831

Actions (login required)

View Item