ANALISIS SENTIMEN MENGGUNAKAN METODE SUPPORT VECTOR MACHINE BERDASARKAN ULASAN PENGGUNA TERKAIT FAKTOR-FAKTOR KELEBIHAN DAN KEKURANGAN FITUR APLIKASI SEGARI

Farhan Habib, . (2024) ANALISIS SENTIMEN MENGGUNAKAN METODE SUPPORT VECTOR MACHINE BERDASARKAN ULASAN PENGGUNA TERKAIT FAKTOR-FAKTOR KELEBIHAN DAN KEKURANGAN FITUR APLIKASI SEGARI. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.

[img] Text
ABSTRAK.pdf
Download (11kB)
[img] Text
AWAL.pdf
Download (710kB)
[img] Text
BAB 1.pdf
Download (99kB)
[img] Text
BAB 2.pdf
Download (256kB)
[img] Text
BAB 3.pdf
Download (133kB)
[img] Text
BAB 4.pdf
Download (493kB)
[img] Text
BAB 5.pdf
Download (86kB)
[img] Text
DAFTAR PUSTAKA.pdf
Download (156kB)
[img] Text
RIWAYAT HIDUP.pdf
Download (106kB)
[img] Text
LAMPIRAN.pdf
Download (1MB)
[img] Text
HASIL PLAGIARISME.pdf
Restricted to Repository staff only
Download (312kB)
[img] Text
ARTIKEL KI.pdf
Restricted to Repository staff only
Download (762kB)

Abstract

The background of this research is that many e-groceries companies have difficulty surviving in the post-pandemic market. Obvious examples are HappyFresh which discontinued its services due to financial difficulties and TaniHub which closed its services due to difficulties maintaining profitability. However, Segari managed to survive and experience positive business growth. This research aims to understand the factors that make the Segari application able to survive until now, identify its advantages and disadvantages, and measure how much user sentiment towards using the application through sentiment analysis. The benefit is knowing the factors that are favored and which need to be improved on the Segari application. The research method used is the support vector machine and linear kernel classification method with 5437 Segari review data from the Google Play Store and App Store. The results showed that positive class comments were dominant in Segari application reviews, as much as 5107 of the total data and the remaining negative class as much as 330 review data. The conclusion of this study is that the most frequently discussed topics for positive sentiments include product freshness, speed of the ordering and delivery process, attractive promos, and high quality, while negative sentiments include technical and service issues, shipping and couriers, and user experience. It is suggested that Segari can improve its services from the negative sentiments obtained. The performance of the model created using the support vector machine method assisted by a linear kernel obtained an accuracy result of 96 percent.

Item Type: Thesis (Skripsi)
Additional Information: [No.Panggil: 2010512124] [Pembimbing 1: Tri Rahayu] [Pembimbing 2: Muhammad Panji Muslim] [Penguji 1: Ika Nurlaili Isnainiyah] [Penguji 2: Rudhy Ho Purabaya]
Uncontrolled Keywords: Sentimen Analysis, Support Vector Machine, Kernel Linear, Segari, E-groceries
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Q Science > QA Mathematics > QA76 Computer software
Divisions: Fakultas Ilmu Komputer > Program Studi Sistem Informasi (S1)
Depositing User: FARHAN HABIB
Date Deposited: 12 Sep 2024 04:40
Last Modified: 12 Sep 2024 04:40
URI: http://repository.upnvj.ac.id/id/eprint/31517

Actions (login required)

View Item View Item