Journal of Applied Data Sciences
Vol 4, No 4: DECEMBER 2023

Text Mining Application With K-Means Clustering to Identify Sentiments and Popular Topics: A Case Study of The Three Largest Online Marketplaces in Indonesia

Andree E Widjaja (Department of Information System, Universitas Pelita Harapan, Tangerang, Indonesia)
Andy Fransisko (Department of Information System, Universitas Pelita Harapan, Tangerang, Indonesia)
Calandra Alencia Haryani (Department of Information System, Universitas Pelita Harapan, Tangerang, Indonesia)
Hery Hery (Department of Information System, Universitas Pelita Harapan, Tangerang, Indonesia)



Article Info

Publish Date
06 Dec 2023

Abstract

The number of internets and social media users, which continues to increase at a very fast rate, has resulted in the emergence of new business opportunities in Indonesia. One of those indications is the emergence of marketplace companies in Indonesia. The presence of these online marketplaces provides people with more online marketplace choices according to their preferences. One of the factors that became the basis for this election was reading comments or reviews from consumers on the marketplace posted on social media. This research was conducted using text mining method with k-means clustering algorithm to systematically identify the sentiments and topics that are widely discussed by online marketplace consumers in Indonesia. The data was processed by the k-means algorithm in the form of comments or reviews from three online marketplaces (Tokopedia, Shopee and Bukalapak) on Twitter. The amount of data for each marketplace referred to was 1500 data tweets. The results showed that the three online marketplaces were associated to different topics, even though they are in the same industry. These differences arise due to the fact that most consumers discuss the topics of programs held by their respective online marketplaces. The main topics related to Tokopedia are “belanja” (“shopping”) and “terimakasih” (“thank you”); while for Shopee “pilih” (“choose”) and “jongho”, and for Bukalapak “pra-kerja” (“pre-employment”). In addition, the sentiment analysis carried out shows that the sentiment of the three online marketplaces is predominantly neutral.

Copyrights © 2023






Journal Info

Abbrev

JADS

Publisher

Subject

Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management

Description

One of the current hot topics in science is data: how can datasets be used in scientific and scholarly research in a more reliable, citable and accountable way? Data is of paramount importance to scientific progress, yet most research data remains private. Enhancing the transparency of the processes ...