A novel approach for website aesthetic evaluation based on convolutional neural networks
Masoud Ganj Khani, Mohammad Reza Mazinani, Mohsen Fayyaz, Mojtaba Hoseini · 2016
In this paper we propose a website aesthetic evaluation method. For achieving better performance, we have applied convolutional neural networks, which are one of the methods of deep learning research area. Using deep learning and convolutional neural networks for feature representation is one of the main tips that makes difference between our work and previous ones. Our system takes a screenshot of the website as input, and finally reports it is a good or bad website based on users' country or not. For evaluation process, we represent the website screenshot using MemNet convolutional neural network. Then we decrease the extracted features dimension using principal component analysis algorithm. Finally, we classify them using a SVM classifier, which trained, based on users' ratings. Furthermore, aesthetics evaluation in this research is language independent. It means the website's language is not important and our method works for all languages.