Mining Users’ Reviews to Support the Release Planning of Mobile Apps (2015-2017)

The mobile apps market is a tremendous success, with millions of apps downloaded and used every day by users spread all around the world. For apps’ developers, having their apps published on one of the major app stores (e.g., Google Play market) is just the beginning of the apps lifecycle. Indeed, in order to successfully compete with the other apps in the market, an app has to be updated frequently by adding new attractive features and by fixing existing bugs. Clearly, any developer interested in increasing the success of her app should try to implement features desired by the app’s users and to fix bugs affecting the user experience of many of them. A precious source of information to decide how to collect users’ opinions and wishes is represented by the reviews left by users on the store from which they downloaded the app, but to exploit such information the app’s developer should manually read each user review and verify if it contains useful information (e.g., suggestions for new features). This is something not doable if the app receives hundreds of reviews per day, as happens for the very popular apps on the market.

In this work, we aim at providing support to mobile apps developers by developing a novel approach exploiting data mining, natural language processing, machine learning, and clustering techniques in order to classify the user reviews on the basis of the information they contain (i.e., useless, suggestion for new features, bugs reporting). Such an approach has been made available in a web-based platform publicly available available to all apps’ developers.

Nowadays, the app market paradigm is the dominating form to distribute a mobile application, with millions of apps downloaded and used every day by users spread all around the world. For developers of mobile applications, having their apps published on one of the major app stores (e.g., Google Play market) is just the beginning of the apps lifecycle. Indeed, to successfully compete with the other apps in the market, an app has to be updated frequently by adding new attractive features, adapting to the developments of hardware, fixing defects, and so on. In fact, any developer interested in increasing the success of her app should try to constantly increase the user experience. A precious source of information to understand users’ opinions and wishes are the reviews left by users on the store from which they downloaded the app. Currently, to exploit such information the app’s developer should manually read each user review and verify if it contains useful information (e.g., suggestions for new features). This is not feasible if the app receives hundreds of reviews per day, as happens for the very popular apps on the market.

In this work, we aim at providing support to mobile apps developers by developing techniques and tools exploiting data mining, natural language processing, machine learning, and clustering algorithms to classify the users’ reviews on the basis of the information they contain (i.e., not relevant, suggestion for new features, defect reporting). The automatic classification of the user reviews can be exploited by the apps’ developers to take conscious decisions on how to evolve their apps (e.g., which features to implement, which defects to fix first, etc).

Involved researchers: Gabriele Bavota, Barbara Russo, Andrea Janes

Contact person for the Free University of Bozen-Bolzano: Gabriele Bavota