A systematic literature review: Opinion mining studies from mobile app store user reviews

Abstract

The mobile app ecosystem and user reviews contain a wealth of information about user experience and expectations.Developers and app store regulators could leverage the information to better understand their audience.The automated extraction of app features in online reviews does not consider the nature of the review text.Opinion spam or fake review detection is one of the largest problems in the domain. As mobile devices have overtaken fixed Internet access, mobile applications and distribution platforms have gained in importance. App stores enable users to search for, purchase and install mobile applications and then give feedback in the form of reviews and ratings. A review might contain information about the users experience with the app and opinion of it, feature requests and bug reports. Hence, reviews are valuable not only to users who would like to find out what others think about an app, but also to developers and software companies interested in customer feedback.The rapid increase in the number of applications and total app store revenue has accelerated app store data mining and opinion aggregation studies. While development companies and app store regulators have pursued upfront opinion mining studies for business intelligence and marketing purposes, research interest into app ecosystem and user reviews is relatively new. In addition to studies examining online product reviews, there are now some academic studies focused on mobile app stores and user reviews.The objectives of this systematic literature review are to identify proposed solutions for mining online opinions in app store user reviews, challenges and unsolved problems in the domain, any new contributions to software requirements evolution and future research direction.