A Fuzzy-Rule-Based Approach for Webpage Aesthetics Modeling

A beautiful webpage design has proven to be one of the decisive factors in user satisfaction. The relationships between design features and webpage aesthetics are often explained by using linear regression models. However, these models are unable to depict the fuzziness and non-linearity of human perceptions. To address these limitations, this study proposes a fuzzy-rule-based model to correlate webpage design features with user ratings on webpage aesthetics. This model gains higher predictive performance with fewer variables in comparison with a linear regression model based on the same experimental data, and classifies webpages and yields more specific design knowledge.