Beyond the Stars: Towards a Novel Sentiment Rating to Evaluate Applications in Web Stores of Mobile Apps

This paper proposes an approach to evaluate mobile applications which complements the information provided by the number of stars and downloads in app stores. The goal is to provide novel information to assist users in the decision-making process regarding the choice of applications. In this sense, we conducted experiments to verify the relationship between the number of stars and the content of review comments. Results indicated that there is information in reviews not properly represented by stars. Thus, we present a sentiment rating generated automatically by aggregating opinions reported in the reviews related to each application. We evaluated this new rating using 26,996 reviews related to six applications present on the Google Play Store. The obtained results allow us to demonstrate that: (1) it is possible and feasible to generate a sentiment rating automatically and (2) the rating is useful for web stores of mobile applications to improve their mechanisms of ranking and recommendation as well as to assist users and developers to evaluate the quality and/or acceptance of the offered mobile applications.