Subagyo, Adinda Rizka Chlorine (2021) Analisis Sentimen Untuk Menentukan Popularitas Marketplace Menggunakan Metode Naive Bayes Classifier. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Text
10611710000051-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (1MB) | Request a copy |
Abstract
Pada bulan Maret 2019, Indonesia berada di peringkat ke-5 dunia
dalam jumlah pengguna internet terbanyak yaitu 143,26 juta pengguna.
Salah satu situs yang sering dikunjungi oleh pengguna internet di
Indonesia yaitu marketplace. Marketplace adalah wadah komunitas bisnis
interaktif secara elektronik yang menyediakan pasar di mana perusahaan
dapat ambil andil dalam e-commerce dan atau kegiatan e-business lainnya
Jumlah marketplace di Indonesia dari tahun ke tahun semakin bertambah.
Hal ini mengakibatkan masyarakat kebingungan untuk memilih situs
belanja online dengan proses transaksi yang aman dan nyaman. Tiga
marketplace tertinggi yang memiliki reputasi baik di Indonesia yaitu
Shopee, Tokopedia dan Lazada. Oleh karena itu, dilakukan penelitian
untuk menentukan popularitas marketplace di antara Shopee, Tokopedia
dan Lazada menggunakan analisis sentimen. Sentimen merupakan
penilaian atau pendapat mengenai layanan pada marketplace, yang
menangkap pendapat positif maupun negatif. Suatu marketplace yang
banyak mendapatkan respon positif, menunjukkan bahwa marketplace
tersebut memiliki popularitas tinggi dibandingkan marketplace lainnya.
Penelitian ini menggunakan metode Naïve Bayes Classifier, dengan
sumber data ulasan Shopee, Tokopedia dan Lazada yang diperoleh dari
situs Google Play. Hasil dari penelitian ini didapatkan Lazada memimpin
sebagai marketplace yang memiliki popularitas tertinggi dibandingkan
Shopee dan Tokopedia, dengan peresentase sentimen positif Lazada
sebesar 75%, sedangkan Tokopedia sebesar 70%, dan Shopee sebesar
63%.
=====================================================================================================
In March 2019, Indonesia was ranked 5th in the world for the
highest number of internet users, namely 143.26 million users. One of the
sites frequently visited by internet users in Indonesia is the marketplace.
Marketplace is an interactive electronic business platform that provides
a market where companies can take part in e-commerce and / or other ebusiness activities. The number of marketplaces in Indonesia is increasing
from year to year. This has resulted in confusion for the public to choose
an online shopping site with a safe and comfortable transaction process.
The three highest reputable marketplaces in Indonesia are Shopee,
Tokopedia and Lazada. Therefore, a study was conducted to determine
the popularity of the marketplace among Shopee, Tokopedia and Lazada
using sentiment analysis. Sentiment is an assessment or opinion about
services on the marketplace, which captures positive or negative
opinions. A marketplace that gets a lot of positive responses, shows that
the marketplace has high popularity compared to other marketplaces.
This study uses the Naïve Bayes Classifier method, with data sources for
Shopee, Tokopedia and Lazada reviews obtained from the Google Play
website. The result of this research is that Lazada leads the way as a
marketplace that has the highest popularity compared to Shopee and
Tokopedia, with a percentage of Lazada's positive sentiment of 75%,
while Tokopedia is 70%, and Shopee is 63%.
Item Type: | Thesis (Undergraduate) |
---|---|
Uncontrolled Keywords: | Analisis Sentimen, Marketplace, Naïve Bayes Classifier ============================================================ Marketplace, Naïve Bayes Classifier, Sentiment Analysis |
Subjects: | Q Science Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) Q Science > QA Mathematics > QA9.58 Algorithms |
Divisions: | Faculty of Vocational > 49501-Business Statistics |
Depositing User: | Adinda Rizka Chlorine Subagyo |
Date Deposited: | 14 Aug 2021 02:12 |
Last Modified: | 14 Aug 2021 02:12 |
URI: | http://repository.its.ac.id/id/eprint/86335 |
Actions (login required)
View Item |