ANALISIS SENTIMEN REVIEW KOMENTAR PADA APLIKASI GOJEK DI SITUS GOOGLE PLAY DENGAN MENGGUNAKAN METODE SUPPORT VECTOR MACHINE (SVM) DAN PARTICLE SWARM OPTIMIZATION (PSO)

Kusumawardhana, Gilang (2020) ANALISIS SENTIMEN REVIEW KOMENTAR PADA APLIKASI GOJEK DI SITUS GOOGLE PLAY DENGAN MENGGUNAKAN METODE SUPPORT VECTOR MACHINE (SVM) DAN PARTICLE SWARM OPTIMIZATION (PSO). Other thesis, Universitas Amikom Purwokerto.

[img] Text
COVER.pdf

Download (470kB)
[img] Text
DAFTAR ISI.pdf

Download (627kB)
[img] Text
ABSTRAK.pdf

Download (583kB)
[img] Image
BAB I.pdf
Restricted to Registered users only

Download (357kB)
[img] Image
BAB II.pdf
Restricted to Registered users only

Download (1MB)
[img] Image
BAB III.pdf
Restricted to Registered users only

Download (432kB)
[img] Image
BAB IV.pdf
Restricted to Registered users only

Download (1MB)
[img] Image
BAB V.pdf
Restricted to Registered users only

Download (442kB)
[img] Image
DAFTAR PUSTAKA.pdf
Restricted to Registered users only

Download (633kB)
[img] Text
LAMPIRAN.pdf
Restricted to Repository staff only

Download (1MB)

Abstract

The growth of online transportation market share in Indonesia experienced a huge growth of 900 million US $ in 2015 and in 2018 it reached 3.8 billion US $ based on Google & TAMASEK 2018. Gojek is one of the most popular online transportation services in Indonesia. On the Google Play site at the end of September 2019, Gojek was recorded to have been downloaded more than 50 million times and has a rating of 4.5 with 2,753,175 comments. While Grab, Gojek's toughest competitor has better statistics with more than 100 million downloads and has a rating of 4.7 with 4,692,497 comments. These comments range from positive to negative. By analyzing comments, the company can understand the shortcomings of the application and the expectations of its users. The purpose of this research is to do sentiment analysis using review data on the google play site to find out what is often reviewed by users and the results can be used for evaluation by Gojek to improve the quality of service. By using the classification method Support Vector Machine and Particle Swarm Optimization to classify reviews into positive and negative sentiment classes. Then the information obtained is visualized using a chart. The results of the analysis using Support Vector Machine and Particle Swarm Optimization produce the best accuracy of 73.40%, an increase of 8.8% before using Particle Swarm Optimization which is 64.60%. The most frequently reviewed positive reviews are "gopay", while the most frequently reviewed negative reviews are "drivers".

Item Type: Thesis (Other)
Additional Information: Dosen Pembimbing: Rizki Wahyudi, M.Kom., dan Fiby Nur Afiana, S.Kom., M.MSI.
Uncontrolled Keywords: Particle Swarm Optimization, Sentiment Analysis, Support Vector Machine, Text Mining.
Subjects: T Technology > T Technology (General)
Divisions: Fakultas Ilmu Komputer > Sistem Informasi
Depositing User: UPT Perpustakaan Pusat Universitas Amikom Purwokerto
Date Deposited: 22 Mar 2021 06:19
Last Modified: 22 Mar 2021 06:19
URI: http://eprints.amikompurwokerto.ac.id/id/eprint/734

Actions (login required)

View Item View Item