Researchers Analyze Age-Related Concerns in Mobile App Reviews Using Machine Learning

Global: Researchers Analyze Age-Related Concerns in Mobile App Reviews Using Machine Learning

Researchers have examined how users of different ages discuss mobile applications by analyzing a curated set of reviews from the Google Play Store. The study, conducted by a team of computer scientists, collected 4,163 reviews, identified 1,429 as age‑related, and applied a suite of eight machine‑learning and large‑language‑model techniques to automatically detect those discussions. The work aims to inform developers about age‑specific challenges and improve the inclusivity of mobile apps.

Dataset Construction and Annotation

The authors manually reviewed each of the 4,163 Google Play entries, separating them into age‑related (1,429) and non‑age‑related (2,734) categories. This labeling process provided a balanced benchmark for training and evaluating classification models, ensuring that the nuances of age‑specific language were captured accurately.

Model Evaluation Across Multiple Techniques

Eight distinct algorithms—including traditional machine‑learning classifiers, deep‑learning architectures, and large language models—were trained on the annotated dataset. Performance metrics such as precision, recall, and F1‑score were calculated to compare effectiveness in detecting age‑related content.

RoBERTa Achieves Highest Precision

Among the evaluated models, the RoBERTa transformer achieved the highest precision, recording a score of 92.46%. This result indicates that RoBERTa was particularly adept at correctly identifying reviews that contain age‑related discussions, minimizing false positives.

Qualitative Insights from Age‑Related Reviews

A subsequent qualitative analysis of the 1,429 age‑related reviews revealed six dominant themes, ranging from concerns about inappropriate content for younger users to usability challenges faced by older adults with vision or cognitive impairments. These themes highlight specific pain points that developers may address through design adjustments or content filtering.

Implications for Mobile App Development

The findings suggest that incorporating automated age‑discussion detection into app‑store analytics could enable developers to proactively identify and remediate age‑specific issues. By tailoring interfaces, accessibility options, and content moderation policies, creators can enhance the overall user experience across the lifespan.

Limitations and Directions for Future Research

The study relies on a single platform (Google Play) and a snapshot of reviews, which may limit generalizability to other ecosystems or time periods. Future work could expand the dataset to include multiple app stores, longitudinal data, and cross‑cultural perspectives to deepen understanding of age‑related user needs.

This report is based on information from arXiv, licensed under Academic Preprint / Open Access. Based on the abstract of the research paper. Full text available via ArXiv.