Published by LPMP Imperium Journal homepage: https://ejournal.
org/index.
php/SERAMBI Perception of user opinions towards sharia mobile banking applications in Indonesia based on comments on Google Play
Store
Reza Edi Saputra3
SERAMBI
1 Sharia Economics.
Faculty of Economics and Business.
Universitas Cholisa Rosanti1.
Fenilinas Adi Artanto Muhammadiyah Pekajangan Pekalongan.
Indonesia 2,3 Informatics.
Faculty of Engineering and Computer Science.
Universitas Muhammadiyah Pekajangan Pekalongan.
Indonesia Abstract This research aims to evaluate user perceptions of the performance of m-banking applications at Islamic banks by analyzing user reviews on the Google Play Store from 4 July 2023 to 4 July 2024.
The research results show that the BSI Mobile and Jago Syariah applications are the most downloaded and have the most reviews, indicating a high number of active users.
Analysis using the Bag of Words (BoW) and Support Vector Regression (SVR) methods shows excellent performance in predicting user sentiment, with an accuracy of more than 90% on six sharia m-banking applications, where the Aladin Syariah application achieved the highest accuracy of 98.
Research findings show that positive perceptions of the app's ease of use contribute to user loyalty.
At the same time, complaints regarding the login and registration process must be improved.
Positive reviews such as "good job" and "steady" dominate the BSI Mobile and Jago Syariah applications, reflecting the popularity of both applications.
ARTICLE INFO
Received 12/11/2024 Revised: 12/1/2024 Accepted: 12/12/2024 Online First 1/16/2025 Public interest statement The unique features such as gold transactions on the Pegadaian Syariah application provide competitive advantages that other Sharia banks can adopt.
These results provide strategic insights for Islamic banks to improve and develop their m-banking services.
Keywords: Application.
Sharia Bank.
Google Play Store.
Sentiment:
Paper type: Brief report Corresponding:
Fenilinas Adi Artanto.
Email: fenilinasadi@gmail.
A The Author.
2025 This work is licensed under a Creative Commons Attribution 4.
International License.
SERAMBI: Jurnal Ekonomi Manajemen dan Bisnis Islam.
Vol 7.
No.
1, 2025, 97-108
eISSN 2685-9904 DOI: https://doi.
org/10.
36407/serambi.
Published by LPMP Imperium Abstrak Penelitian ini bertujuan untuk mengevaluasi persepsi pengguna terhadap kinerja aplikasi mbanking pada bank syariah melalui analisis ulasan pengguna di Google Play Store selama periode 4 Juli 2023 hingga 4 Juli 2024.
Hasil penelitian menunjukkan bahwa aplikasi BSI Mobile dan Jago Syariah merupakan yang paling banyak diunduh dan memiliki ulasan terbanyak, mengindikasikan tingginya jumlah pengguna aktif.
Analisis menggunakan metode Bag of Words (BoW) dan Support Vector Regression (SVR) menunjukkan kinerja yang sangat baik dalam memprediksi sentimen pengguna, dengan akurasi lebih dari 90% pada enam aplikasi m-banking syariah, dimana aplikasi Aladin Syariah mencapai akurasi tertinggi sebesar 98,54%.
Temuan penelitian menunjukkan bahwa persepsi positif terhadap kemudahan penggunaan aplikasi berkontribusi pada loyalitas pengguna, sementara keluhan terkait proses login dan pendaftaran perlu diperbaiki.
Ulasan positif seperti "good job" dan "mantap" mendominasi pada aplikasi BSI Mobile dan Jago Syariah, mencerminkan popularitas kedua aplikasi tersebut.
Pernyataan kepentingan publik Ffitur unik seperti transaksi emas pada aplikasi Pegadaian Syariah memberikan keunggulan kompetitif yang dapat diadopsi oleh bank syariah lainnya.
Hasil ini memberikan wawasan strategis bagi bank syariah untuk meningkatkan dan mengembangkan layanan m-banking Kata Kunci: Aplikasi: Bank Syariah.
Google Playstore.
Sentiment Introduction With the development of information and communication technology, banking services are shifting to digital platforms, including in the Sharia banking sector.
Islamic banks now provide banking services that comply with Sharia principles and strive to improve user experience by developing mobile banking .
-bankin.
applications that are easy to use and reliable.
maintain customer loyalty who have used m-banking applications.
Islamic bank management needs to understand user perceptions and satisfaction with the performance of the applications they provide (Anggraeni et al.
, 2.
To improve quality and user experience.
Islamic banks need to understand customer perceptions of the performance of their m-banking applications.
This perception includes ease of use, voice transmission, security, and speed in processing user transactions (Arifiyanti et , 2.
One way to understand user perceptions is to analyze review data from platforms like the Google Play Store or social media.
This data management review offers valuable insights into user preferences and feedback, which can be used to develop service quality improvement strategies (Subowo et al.
, 2.
In this context.
Islamic bank management can use sentiment from m-banking application user reviews to see whether their opinions tend to be more positive, neutral, or negative regarding various aspects of the application, such as ease of use, transaction speed, and security.
SERAMBI, 7.
, 97-108
Cholisa Rosanti, et al.
Application.
Sharia Bank To manage the perceptions of Islamic bank customers who use m-banking, they can use sentiment analysis methods to identify areas that need improvement.
Sentiment analysis is a text-mining method that classifies data as text (Fatkhudin et al.
, 2.
One of the methods in sentiment analysis is Neural Bag of Words (NBOW) and Support Vector Regression (SVR), which provide a simple but effective approach to modeling user review data without requiring complex computations (Polap & Wlodarczyk-Sielicka, 2.
The NBOW method allows review data to be converted into vector representation for further analysis, where text data in the form of review sentences will be interpreted into words that will be given weighted values according to the preferences of application reviews according to positive, neutral and negative sentiments (Sheikh et al.
, 2.
Meanwhile.
SVR is used to assess the accuracy of the NBOW weighting classification, which has given positive, neutral, and negative weights to M-Banking Syariah application reviews (Isnaeni et al.
, 2.
Research by Haq et al.
entitled Sentiment Analysis of Livin Application Reviews on Google PlayStore shows that the majority of Livin application users give positive reviews of the Livin application, where positive comments contain reviews with the words easy, transaction, good, pleasing and Independent.
Meanwhile, negative reviews on the Livin application include updates, failures, and errors.
Sentiment analysis in the Livin application provides input to application developers and Bank Mandiri in providing optimal customer It is focused on negative sentiment, which shows that the Livin application has shortcomings regarding updates from the application, which also often experiences failures in The hope is that by correcting errors or using negative reviews as input.
Mandiri Bank's services will improve.
Rizky Pratama et al.
analyzed two applications: BRImo and BCA Mobile.
This research obtained satisfaction results using the BCA Mobile application of 95% and only 94% for the BRImo application.
The Brimo application shows that most users mention difficult This word shows the difficulty of registering on the BRImo application.
Meanwhile.
BCA Mobile displays the word verification, which means there is difficulty in verifying the Rachmawati Oktaria Mardiyanto et al.
researched sharia m-banking applications only on the Bank Syariah Indonesia (BSI) application.
This research shows that most users gave positive reviews, mentioning the words cool and thank God, while negative reviews mentioned the word login.
this shows that customers experienced problems logging into the Different from previous research, this research focuses on Sharia banks, where there are six m-banking applications from Sharia banks, namely Bank Jago Syariah.
Bank Aladin Syariah.
Pegadaian Syariah.
Bank Syariah Indonesia (BSI Mobil.
BCA Syariah, and Mega Syariah which will be tested and compared so that later it will also show which Islamic banking application is superior.
This research aims to evaluate the perceptions of sharia mbanking application users' sentiments towards the sharia m-banking application so that the results of this analysis can later be used to improve the application's performance and increase user satisfaction.
The results of this research can be a reference for bank management in improving their m-banking applications so that they are more responsive to the needs of their users and can provide increased user satisfaction and loyalty on an ongoing basis.
DOI: https://doi.
org/10.
36407/serambi.
Published by LPMP Imperium Methods This research uses a quantitative approach based on secondary data analysis to evaluate users' perceptions of the Sharia m-banking application through comment reviews on Google Playstore.
The data taken are comments from reviews of m-banking applications, namely the Bank Jago Syariah.
Bank Aladin Syariah.
Pegadaian Syariah.
Bank Syariah Indonesia (BSI Mobil.
BCA Syariah, and Mega Syariah applications.
Data was taken from July 4, 2023, to July 4, 2024, with a data period of 1 year.
Research stages include data collection, data processing through feature extraction, data separation for analysis, and evaluation of results.
All stages are carried out with the help of the Machine Learning method with the Python programming language, which is carried out using Google Collabs.
Data collection.
In this stage, user review data is collected from the Google Play Store.
the data collection method uses the Google Play Store API with a data crawling method using the Python programming language (Eskiyaturrofikoh & Suryono, 2.
The data collected includes reviews in the form of comments.
Feature extraction and data separation.
Feature extraction aims to represent quantitative data from the review text, which will later be analyzed (Muttaqin & Kharisudin.
The Bag of Words (BoW) method converts text data into a feature matrix that will show the frequency of occurrence of words from each comment review in the form of text (Pohan et , 2.
This process includes removing irrelevant common words .
top word.
, data normalization, and stemming to simplify word forms (Yoshikawa, 2.
After that, the data is divided into training data and test data with a ratio of 8:2.
the training data is used to build the model, while the test data is used to evaluate the performance of the model (Rahman.
Analysis and evaluation.
In analyzing and evaluating the data, the Support Vector Regression (SVR) model was used to determine the m-banking application's sentiment.
SVR
is used because of its ability to handle high-dimensional data and provide accurate results in sentiment classification (Difitria & Cholissodin, 2.
After the model is trained using training data, evaluation is carried out on test data using the Mean Squared Error (MSE) matrix to see and measure how accurate the predictions that have been made are (Isnaeni et al.
, 2.
Evaluation.
After the support Vector Regression method predicts review sentiment on Google Play Store comments about the M-Mobile Bank Syariah application, the results will be analyzed to determine the tendency of user perceptions towards aspects of the application (Prasetyo & Siahaan, 2.
The findings will be evaluated to provide strategic recommendations to Islamic bank management regarding improving the quality of M-banking This analysis will provide an understanding of dimension management that will influence user satisfaction, so it will provide planning in feature development and improving M-Banking services at Sharia Banks.
Results and discussion The data collection process was carried out using the Google Play Scraper method.
This library can be used in Google Collab.
The data collection period starts from July 4, 2023, to July 4, 2024, where a total of 160,026 comments were obtained on the Google Play Store.
The following is the amount of data and the number of downloads from each Sharia Bank m-mobile SERAMBI, 7.
, 97-108 Cholisa Rosanti, et al.
Application.
Sharia Bank Table 1 Number of reviews and Downloads Application Review Download Jago Syariah > 10 Million Aladin Syariah < 1 Million Pegadaian Syariah < 500 Thousand BSI Mobile > 10 Million BCA Syariah < 500 Thousand Mega Syariah < 100 Thousand Table 1 shows that the most downloaded applications are BSI Mobile and Jago Syariah, and the least downloaded applications are Mega Syariah.
This indirectly shows that there are more users of the BSI Mobile application and that they are using m-banking at Sharia banks, which have many users.
This indirectly also shows that the number of customers from BSI Mobile is more than that of other Sharia banks.
Figure 1 Review Graph Figure 1 and Table 1 show that BSI Mobile receives the highest number of reviews, whereas Mega Syariah has the fewest.
This trend corresponds with the number of downloads of the M-Banking Syariah applications, where the most downloaded applications also tend to have more reviews.
This data indicates that a higher number of users typically leads to increased reviews for an application.
Once the review data is collected, the next step is to label the reviews based on their star ratings.
Specifically, scores below two are classified as unfavorable, three are considered DOI: https://doi.
org/10.
36407/serambi.
Published by LPMP Imperium neutral, and scores above four are classified as positive.
A preprocessing method is applied to remove irrelevant words, such as hashtags and URLs.
After preprocessing, weighting is performed, which results in an accuracy level for the sharia m-banking application as Table 2 Method Accuracy Application Review Accuracy Jago Syariah Aladin Syariah Pegadaian Syariah BSI Mobile BCA Syariah Mega Syariah Table 2 demonstrates that the Aladin Syariah application achieved the highest accuracy using the Bags of Words (BoW) method.
In contrast, the Mega Syariah application had the lowest These results indicate that the BoW method is effective for sentiment analysis of reviews for Sharia banking mobile applications, as it consistently yields accuracy rates above 90% across all assessed applications.
In contrast, research conducted by Polap and Wlodarczyk-Sielicka .
, which utilized a CNN method, attained an accuracy of only 84%.
Additionally, words from the reviews were organized using a word cloud, and the resulting findings were as follows:
Jago Syariah Figure 2 shows the results of Wordcloud on the Jago Syariah application.
It can be seen that the positive reviews contain the words excellent, fast, helpful, and like.
the neutral reviews show the words please and login, while the negative reviews contain the words register, login, and account.
This shows that the users of the Jago Syariah application feel stable.
The application is helpful and easy, but what needs to be improved about the application can be seen in the negative reviews about login and accounts, where users find it challenging to create an account and the login process on the Jago Syariah application.
Figure 2 Wordcloud Jago Syariah SERAMBI, 7.
, 97-108
Cholisa Rosanti, et al.
Application.
Sharia Bank Aladin Syariah Figure 3 displays the results of a word cloud generated from reviews of the Aladin Syariah The analysis reveals that positive reviews frequently mention "transfer" and "digital bank," indicating that users appreciate the application's banking features.
In contrast, neutral reviews contain phrases like "please" and "register," while negative reviews highlight challenges with the terms "register," "enter," and "email.
" These results indicate that users perceive Aladin Syariah as a digital banking application primarily used to transfer funds.
However, the negative reviews suggest areas for improvement, particularly regarding the registration and login processes, as users report difficulties creating accounts and accessing the application.
Figure 3.
Wordcloud Aladin Syariah Pegadian Syariah Figure 4 presents the results of the word cloud analysis for the Pegadaian Syariah application.
The positive reviews highlight "helpful" and "good," indicating that users generally view the Sharia pawnshop as beneficial.
The neutral reviews feature "gold" and "bank," reflecting the app's primary functions.
In contrast, the negative reviews mention "login," "register," and "update," suggesting areas that require improvement.
Users report difficulties with the account registration process and logging into the application.
Additionally, there are concerns about the update process within the sharia pawnshop app.
Figure 3 Wordcloud Pegadaian Syariah DOI: https://doi.
org/10.
36407/serambi.
Published by LPMP Imperium BSI Mobile Figure 5 displays the results of a word cloud analysis of the BSI Mobile application.
The positive reviews highlight phrases such as "complete features" and "good job," indicating that users are satisfied with the application's offerings.
In contrast, the neutral reviews focus on the word "transaction," while the negative reviews mention terms like "application," "complicated," and "error.
" This result suggests that while users appreciate the comprehensive features and express satisfaction with the service, areas need improvement.
Specifically, users find the application complex and prone to errors, illustrating a demand for a more userfriendly experience.
Figure 4 Wordcloud BSI Mobile BCA Syariah Figure 6 illustrates the results of a word cloud analysis for the BCA Syariah application.
The positive reviews highlight words such as "easy," "good," and "helpful," indicating that users appreciate the application's features.
In contrast, the neutral reviews contain terms like "transaction," "transfer," and "features," suggesting a focus on the application's functionality.
Negative reviews primarily mention issues related to "log in" and "register," indicating that users experience difficulties accessing the application or creating accounts.
Overall, while users find the BCA Syariah application helpful and its features beneficial, improvements are needed in the login and registration processes.
Figure 5 Wordcloud BCA Syariah SERAMBI, 7.
, 97-108
Cholisa Rosanti, et al.
Application.
Sharia Bank Mega Syariah Figure 7 illustrates the word cloud results for the Mega Syariah application.
The analysis reveals that positive reviews frequently use "helpful" and "easy.
" In contrast, neutral reviews mention "help" and "SMS," while negative reviews highlight "login," "register," and "difficult.
" These results suggest that users generally find the Mega Syariah application helpful and easy to navigate.
However, the negative reviews indicate areas for improvement, particularly regarding the account registration and login processes, which users have described as challenging.
Figure 6 Wordcloud Mega Syariah Table 3 demonstrates that Sharia mobile banking applications with positive reviews are those that are user-friendly and facilitate transactions.
When their experience is positive, users tend to express their appreciation through favorable and solid reviews.
Conversely, most applications receiving negative reviews face criticism due to complicated login and registration processes.
Users expect all applications to provide a straightforward login and registration experience.
Table 3 Words that frequently appear in application reviews Frequently Words Application Positive Netral Negative Jago Syariah Mantab Login Daftar Masuk Aladin Syariah Transfer Bank Saldo Masuk Pegadaian Syariah Bagus.
Mudah Emas Gagal.
Login BSI Mobile Good Job Aplikasi Ribet BCA Syariah Fitur.
Mudah Aplikasi Login Mega Syariah Membantu SMS Daftar Discussion The analysis conducted using the Bags of Words (BoW) and Support Vector Regression (SVR) methods indicates that these techniques are effective for sentiment analysis in Islamic banks.
DOI: https://doi.
org/10.
36407/serambi.
Published by LPMP Imperium The findings reveal that the M-Mobile application at Sharia Bank achieved an impressive overall accuracy of over 90%.
Specifically, the Aladin Syariah application attained the highest accuracy at 98.
In contrast, although slightly lower, the Mega Syariah application still reported a commendable accuracy of 98.
The review collection from users of m-banking applications at Islamic banks on the Google Play Store emphasizes the importance of maintaining positive reviews to retain application users.
Users appreciate the application's ease of use while addressing negative reviews, which is crucial for application development.
Many users have expressed difficulties with the login and registration processes, indicating a strong desire to improve these areas.
Among the user reviews, the m-mobile banking applications at Sharia Banks received considerable praise, highlighting excellent functionality and reliability.
The BSI Mobile application received numerous positive reviews for its performance.
In contrast, the Jago Syariah application was lauded with descriptions such as "great.
" These two applications have gained substantial popularity and many downloads, serving as models for other Sharia banks aiming to enhance their M-Mobile banking applications.
Additionally, the Sharia Pawnshop application stands out for its unique features compared to other applications.
Notably, within neutral sentiment, the Sharia Pawnshop application received praise for its gold transaction feature, setting it apart from other M-Mobile banking applications at Sharia banks.
Conclusion A review of data collected between July 4, 2023, and July 4, 2024, has identified that BSI Mobile and Jago Syariah are the most downloaded applications.
This indicates that these applications have a large user base, making them the most popular among other Sharia banks.
The number of reviews on the Google Play Store supports this observation, showing that BSI Mobile has more active users than Jago Syariah, as evidenced by the higher number of reviews.
This research focuses on six Sharia-based mobile banking applications: Jago Syariah.
Aladin Syariah.
Pegadaian Syariah.
BSI Mobile.
BCA Syariah, and Mega Syariah.
The market for Sharia-based banking is growing, with new entrants such as BTN Syariah and Bank Jateng Syariah that have not yet been evaluated.
It is important to note that the data collection process did not involve filtering out spam reviews.
Additionally, digital marketing techniques can be used to manipulate reviews on the Google Play Store by posting positive review spam, a factor that this study does not address.
References