Evaluation of User Interface Design Metrics using Generator of Realistic-Looking Dashboard Samples

aesthetics, dashboard, generator, usability guidelines, user testing

The analysis of user interfaces using quantitative metrics is a straightforward way to quickly measure interface usability and other various design aspects (such as the suitability of page layout or selected colors). Development and evaluation of objective metrics corresponding with user perception, however, usually requires a sufficiently large training set of user interface samples. Finding real user interface samples might not be easy. Therefore, we rather use generated samples. In such case, we need to provide a realistic-looking appearance of samples. This paper describes a workflow of the preparation of such samples. It presents a configurable generator based on the composition of simple widgets according to a predefined model. It also describes a reusable library for simple creation of widgets using capabilities of the JavaScript framework Vue.js. Finally, we demonstrate the applicability of the generator on a generation of dashboard samples which are used to evaluate existing metrics of interface aesthetics and show the possibility of their improvement.