Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Banking on Feedback: Text Analysis of Mobile Banking iOS and Google App Reviews

Yekta Amirkhalili, Ho Yi Wong

TL;DR

This study analyzes mobile banking app reviews from five Canadian banks across iOS and Google Play using text analytics. It applies LDA with TF-IDF to extract topics and tests multiple sentiment classifiers (TextBlob, VADER, MNB, LSTM), selecting LSTM for iOS (82% accuracy) and MNB for Google (77.29%), with manual labeling as a reference. The results reveal recurring issues around login, updates, and app stability, while positive reviews reward usability and customer service, offering actionable guidance for improving user experience. The work demonstrates the value of cross-platform review analysis for informing design, quality assurance, and customer support practices in m-banking.

Abstract

The rapid growth of mobile banking (m-banking), especially after the COVID-19 pandemic, has reshaped the financial sector. This study analyzes consumer reviews of m-banking apps from five major Canadian banks, collected from Google Play and iOS App stores. Sentiment analysis and topic modeling classify reviews as positive, neutral, or negative, highlighting user preferences and areas for improvement. Data pre-processing was performed with NLTK, a Python language processing tool, and topic modeling used Latent Dirichlet Allocation (LDA). Sentiment analysis compared methods, with Long Short-Term Memory (LSTM) achieving 82\% accuracy for iOS reviews and Multinomial Naive Bayes 77\% for Google Play. Positive reviews praised usability, reliability, and features, while negative reviews identified login issues, glitches, and dissatisfaction with updates.This is the first study to analyze both iOS and Google Play m-banking app reviews, offering insights into app strengths and weaknesses. Findings underscore the importance of user-friendly designs, stable updates, and better customer service. Advanced text analytics provide actionable recommendations for improving user satisfaction and experience.

Banking on Feedback: Text Analysis of Mobile Banking iOS and Google App Reviews

TL;DR

This study analyzes mobile banking app reviews from five Canadian banks across iOS and Google Play using text analytics. It applies LDA with TF-IDF to extract topics and tests multiple sentiment classifiers (TextBlob, VADER, MNB, LSTM), selecting LSTM for iOS (82% accuracy) and MNB for Google (77.29%), with manual labeling as a reference. The results reveal recurring issues around login, updates, and app stability, while positive reviews reward usability and customer service, offering actionable guidance for improving user experience. The work demonstrates the value of cross-platform review analysis for informing design, quality assurance, and customer support practices in m-banking.

Abstract

The rapid growth of mobile banking (m-banking), especially after the COVID-19 pandemic, has reshaped the financial sector. This study analyzes consumer reviews of m-banking apps from five major Canadian banks, collected from Google Play and iOS App stores. Sentiment analysis and topic modeling classify reviews as positive, neutral, or negative, highlighting user preferences and areas for improvement. Data pre-processing was performed with NLTK, a Python language processing tool, and topic modeling used Latent Dirichlet Allocation (LDA). Sentiment analysis compared methods, with Long Short-Term Memory (LSTM) achieving 82\% accuracy for iOS reviews and Multinomial Naive Bayes 77\% for Google Play. Positive reviews praised usability, reliability, and features, while negative reviews identified login issues, glitches, and dissatisfaction with updates.This is the first study to analyze both iOS and Google Play m-banking app reviews, offering insights into app strengths and weaknesses. Findings underscore the importance of user-friendly designs, stable updates, and better customer service. Advanced text analytics provide actionable recommendations for improving user satisfaction and experience.

Paper Structure

This paper contains 10 sections, 3 equations, 3 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Data and Methods
  3. Data
  4. Results
  5. Topic Modeling
  6. Sentiment Analysis
  7. Topic Modeling - Sentiment Specific
  8. Negative Reviews Results
  9. Positive Reviews Results
  10. Conclusions

Figures (3)

  • Figure 1: Top most frequent words in iOS and Google corpora by simple count
  • Figure 2: Model performance for different number of topics (iOS data). (left) Perplexity scores, where a lower perplexity score indicates a better model, (right) Coherence scores, where a higher coherence score indicates a better model.
  • Figure 3: Model performance for different number of topics (Google data). (left) Perplexity scores, where a lower perplexity score indicates a better model, (right) Coherence scores, where a higher coherence score indicates a better model.