Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Age Matters: Analyzing Age-Related Discussions in App Reviews

Shashiwadana Nirmania, Garima Sharma, Hourieh Khalajzadeh, Mojtaba Shahin

TL;DR

This study analyzes how age-related concerns are expressed in mobile app reviews to inform age-inclusive design. It builds a manually annotated dataset of 4,163 Google Play reviews (1,429 age-related, 2,734 non-age) and evaluates eight ML/DL/LLM models for automatic detection, with RoBERTa providing the strongest performance. A qualitative thematic analysis of the 1,429 age-related reviews reveals six higher-order themes and 17 sub-themes, covering content suitability, age verification, usability, privacy, interactions, and feature requests. The findings offer actionable guidance for developers to create age-inclusive experiences, including dynamic age restrictions, robust verification, and personalized content while highlighting methodological options for scaling such analyses across software domains. The work demonstrates the potential of combining strong transformer-based classifiers with qualitative insights to surface age-focused issues from large-scale user feedback and suggests avenues for extending this approach to other artifacts in the software lifecycle.

Abstract

In recent years, mobile applications have become indispensable tools for managing various aspects of life. From enhancing productivity to providing personalized entertainment, mobile apps have revolutionized people's daily routines. Despite this rapid growth and popularity, gaps remain in how these apps address the needs of users from different age groups. Users of varying ages face distinct challenges when interacting with mobile apps, from younger users dealing with inappropriate content to older users having difficulty with usability due to age-related vision and cognition impairments. Although there have been initiatives to create age-inclusive apps, a limited understanding of user perspectives on age-related issues may hinder developers from recognizing specific challenges and implementing effective solutions. In this study, we explore age discussions in app reviews to gain insights into how mobile apps should cater to users across different age groups.We manually curated a dataset of 4,163 app reviews from the Google Play Store and identified 1,429 age-related reviews and 2,734 non-age-related reviews. We employed eight machine learning, deep learning, and large language models to automatically detect age discussions, with RoBERTa performing the best, achieving a precision of 92.46%. Additionally, a qualitative analysis of the 1,429 age-related reviews uncovers six dominant themes reflecting user concerns.

Age Matters: Analyzing Age-Related Discussions in App Reviews

TL;DR

This study analyzes how age-related concerns are expressed in mobile app reviews to inform age-inclusive design. It builds a manually annotated dataset of 4,163 Google Play reviews (1,429 age-related, 2,734 non-age) and evaluates eight ML/DL/LLM models for automatic detection, with RoBERTa providing the strongest performance. A qualitative thematic analysis of the 1,429 age-related reviews reveals six higher-order themes and 17 sub-themes, covering content suitability, age verification, usability, privacy, interactions, and feature requests. The findings offer actionable guidance for developers to create age-inclusive experiences, including dynamic age restrictions, robust verification, and personalized content while highlighting methodological options for scaling such analyses across software domains. The work demonstrates the potential of combining strong transformer-based classifiers with qualitative insights to surface age-focused issues from large-scale user feedback and suggests avenues for extending this approach to other artifacts in the software lifecycle.

Abstract

In recent years, mobile applications have become indispensable tools for managing various aspects of life. From enhancing productivity to providing personalized entertainment, mobile apps have revolutionized people's daily routines. Despite this rapid growth and popularity, gaps remain in how these apps address the needs of users from different age groups. Users of varying ages face distinct challenges when interacting with mobile apps, from younger users dealing with inappropriate content to older users having difficulty with usability due to age-related vision and cognition impairments. Although there have been initiatives to create age-inclusive apps, a limited understanding of user perspectives on age-related issues may hinder developers from recognizing specific challenges and implementing effective solutions. In this study, we explore age discussions in app reviews to gain insights into how mobile apps should cater to users across different age groups.We manually curated a dataset of 4,163 app reviews from the Google Play Store and identified 1,429 age-related reviews and 2,734 non-age-related reviews. We employed eight machine learning, deep learning, and large language models to automatically detect age discussions, with RoBERTa performing the best, achieving a precision of 92.46%. Additionally, a qualitative analysis of the 1,429 age-related reviews uncovers six dominant themes reflecting user concerns.
Paper Structure (32 sections, 4 figures, 3 tables)

This paper contains 32 sections, 4 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Age in Software Systems
  4. Human-Centric Discussions in App Reviews
  5. Research Methodology
  6. Dataset
  7. Data Collection
  8. Identifying age-related reviews
  9. Data Annotation
  10. Automatic Detection of Age-related Reviews (RQ1)
  11. Dataset
  12. Models
  13. Traditional ML Models
  14. Pre-trained Language Models
  15. Large Language Models (LLMs)
  16. ...and 17 more sections

Figures (4)

  • Figure 1: The dataset construction process for RQ1 and RQ2
  • Figure 2: GPT Prompt with context
  • Figure 3: t-SNE plot before (left) and after (right) training the best performing RoBERTa model. The plot is created using the test set for the fold corresponding to the best model.
  • Figure 4: The steps of performing thematic analysis.