Citation

Bau, Yoon Teck and Tee, Eng Leong and Goh, Chien Le (2023) Sentiment Analysis of E-Wallet Companies: Exploring Customer Ratings and Perceptions. Journal of Logistics, Informatics and Service Science, 10 (4). pp. 189-205. ISSN 2409-2665

Text 2.pdf - Published Version Restricted to Repository staff only Download (382kB)

Abstract

This sentiment analysis research reports a systematic study of customer reviews of unstructured data for seven popular e-wallet companies, including Alipay, Google Wallet, Grab Superapp, PayPal, Samsung Wallet, Shopee MY, and Touch 'n Go eWallet. Previously, companies faced challenges in effectively utilizing customer reviews to understand and assess customer sentiment toward their products or services. However, with advancements in sentiment analysis techniques, companies can harness the power of customer reviews to gain valuable insights and improve their offerings. The purpose of this study is to explore the use of sentiment analysis in e-wallet companies, where understanding customer sentiment is crucial for enhancing user experiences and driving business success. The research methods employed in this study start by collecting customer reviews spanning four years, from 2019 to 2022. Next, four data pre-processing methods are applied to transform the raw review data into a suitable format for sentiment analysis: data standardization, tokenization, stop word removal, and lemmatization. Sentiment analysis methods were then used to classify reviews as positive, neutral, or negative for each e-wallet company. This study introduced a novel method using rating accuracy to evaluate the polarity sentiment classifications. The results of this study revealed that all e-wallet companies had high rating accuracies for positive sentiments, indicating positive customer sentiment towards their services. However, the rating accuracies for negative sentiments were lower, suggesting challenges in accurately predicting and classifying negative customer sentiment. For neutral sentiments, the rating accuracies were generally low, except for Alipay in 2019, which demonstrated the highest accuracy in capturing customer-neutral sentiment. The evaluation of the findings of this study has important implications for both theory and practice in the e-wallet industry. The practical implications of this study offer concrete guidance for e-wallet companies to enhance their services based on customer sentiments. In contrast, the theoretical implications highlight the need for ongoing research and innovation in sentiment analysis methods that consider other than customer ratings within the e-wallet industry. This study contributes to sentiment analysis by introducing rating accuracy as a measure for evaluating customer reviews accurately. The contribution will provide a more comprehensive understanding of customer sentiments. The methods employed in this study can be applied to enhance sentiment analysis in various customer relationship management domains.

Item Type:	Article
Uncontrolled Keywords:	Sentiment analysis, E-wallet companies, Rating accuracy evaluation, Customer reviews
Subjects:	H Social Sciences > HG Finance > HG1710 Electronic Funds Transfers
Divisions:	Faculty of Computing and Informatics (FCI)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	07 Dec 2023 02:58
Last Modified:	07 Dec 2023 02:58
URII:	http://shdl.mmu.edu.my/id/eprint/11934

Downloads

Edit (login required)