Prioritizing App User Reviews Based on Multi- Dimensional Factors

With the rapid development of the internet, the model of app update and maintenance is shifting from enterprise-driven approaches to collaborative mechanisms powered by real-time user feedback. However, due to the voluminous and heterogeneous nature of user reviews, dynamic user demands, and limited resources, developers urgently require systematic methods to filter high-value feedback. This paper proposes URPM (User Review-Based Prioritization Method), a method for app update and maintenance that prioritizes user reviews through multi-dimensional evaluation to support version optimization decisions. Firstly, the BERTopic topic model is employed to cluster user reviews and calculate topic proportions to quantify user attention. Secondly, sentiment scores of reviews are computed using the SnowNLP sentiment analysis tool. Finally, a priority scoring function integrating user attention, sentiment scores, review timeliness, and user ratings is constructed to rank review priorities. Experimental results demonstrate that the recommendations generated by URPM align with the content adopted by developers, efficiently providing actionable insights for app updates.