| Description |
Dataset of 820 screenshots used during the development of the Appsthetics neural network model for the automatic assessment of Android user interfaces. |
| Data collection |
This dataset is composed of images of screenshots of Android apps developed with App Inventor that are available in the App Inventor Gallery or developed in the initiative Computação na Escola/INCoD/INE/UFSC context. The interfaces were manually selected to avoid duplication of images and to assure ethical aspects. |
| Data formats |
Images in PNG (.png) format with 1080 × 1920 pixels, grouped into a single set, together with three CSV files (.csv): one for training/validation, one for testing and one listing the images removed from training (average absolute deviation greater than .9). Each CSV file contains the image name and the image score per line. The visual aesthetics scores were calculated based on the assessments of ten volunteers. They rated each screenshot on a 5-point semantic differential scale: 1 = very ugly; 2 = ugly; 3 = not ugly and not beautiful; 4 = beautiful; 5 = very beautiful. The final score of each screenshot is the median value of all ratings received converted into the interval [0..1]. |
| Download |
Full dataset |
| License |
Files available are licensed under CreativeCommons Attribution-NonCommercial-ShareAlike 4.0 International. |
| Citation |
Lima, A. L. S., Gresse von Wangenheim, C., (2022). A Deep Learning Model for the Assessment of the Visual Aesthetics of Mobile User Interfaces, GQS/INCoD/INE/UFSC [Data set]. |
| Keywords |
aesthetics; mobile application; Android; deep learning; automatic assessment |
