On the Impact of the User Context Criteria in the Detection of Mobile User Interfaces Defects

Abstract

The evaluation of the mobile user interfaces remains one of the most challenging issues in order to improve the interaction between the user and the mobile applications. In this context, many methods and tools have been proposed to evaluate the quality of Graphical Mobile User Interface (GMUI), like heuristic evaluation and field-testing. However, these available techniques tend to be time consuming and error-prone. An evident solution is to define a set of rules as a combination of metrics quality, context criteria and a list of possible defect types. In fact, the preferences of users are different and heterogeneous. Indeed, the users don't have the same profiles and to satisfy all users' preferences, we can highly decrease the accuracy of the detection rules. In this paper, we consider the generation of the GMUI quality evaluation rules as a multi-objective problem where the goal is to find the best rules maximizing the number of detected aesthetic quality issues and minimizing the complexity of the rules. In addition, we studied the impact of the context criteria in the detection of mobile user defects by applying same feature selection techniques in order to select the most pertinent features of context criteria and to increase the detection rules process.