Webthetics: Quantifying webpage aesthetics with deep learning

Dou, Qi; Zheng, Xianjun Sam; Sun, Tongfang; Heng, Pheng‐Ann

doi:10.1016/j.ijhcs.2018.11.006

Cited by 55 publications

(39 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…According to Fogg (2003), systems that are more attractive are more persuasive. Our ability to determine the aesthetic quality of an object is highly effective, and evaluations form fast, implying a natural tendency of our visual system that is relevant to success (Dou et al 2019). One of the elements of aesthetics is the user impression of visual complexity (Reinecke et al 2013).…”

Section: Mobile Application Characteristicsmentioning

confidence: 99%

Susceptibility to social influence strategies and persuasive system design: exploring the relationship

Halttu

Oinas-Kukkonen

2021

Behaviour & Information Technology

View full text Add to dashboard Cite

Cialdini's six principles of influence are commercially used but less common to encourage healthy behaviors. This study examines how these influence strategies relate to the persuasive systems design (PSD) model constructs implemented in commercial mobile fitness applications. Our research questions address whether susceptibility to influence strategies strengthen the relationships between persuasive constructs and, if so, which strategies and software features are promising to leverage the persuasive effects of systems designed to change health behaviors. This study presents results from a survey of system users (N=147) and their selfreported susceptibility to the six principles. All PSD model constructs showed significantly unequal distribution for some influence strategy, indicating that susceptibility to these strategies affects how users evaluate systems. The commitment principle correlated positively with all persuasive constructs, while reciprocation, scarcity and liking all significantly affected system evaluations. Susceptibility to influence strategies also moderated the relationships between PSD model constructs, but the moderation was often negative with small effect sizes. Our preliminary results indicate that practitioners could benefit from utilizing these influence strategies, especially susceptibility to commitment and reciprocation, which are stable and often high. However, the interaction between these strategies and persuasive systems is not straightforward and would benefit from further research.

show abstract

Section: Mobile Application Characteristicsmentioning

confidence: 99%

Susceptibility to social influence strategies and persuasive system design: exploring the relationship

Halttu

Oinas-Kukkonen

2021

Behaviour & Information Technology

View full text Add to dashboard Cite

show abstract

“…The most common method is subjective surveys [48,50,55,77]. Similar surveys can be used to validate computational models [18,41,67,84,86], including computer vision alternatives to subjective surveys. Higher-level, expertise-and memory-dependent aesthetic components such as familiarity and prototypicality are [25,73].…”

Section: Aesthetic Components and Their Measurementmentioning

confidence: 99%

Aesthetics for Communicative Visualization: a Brief Review

Riahi¹,

Watson²

2021

Preprint

View full text Add to dashboard Cite

We present a brief and somewhat selective review of research on the applied value of aesthetics for communicative visualization. Because aesthetics research in the field of visualization is relatively new, we survey aesthetics research in closely related fields, including psychology, graphic design, and marketing. Because of space limitations, we constrain our discussion of marketing research and omit any examination of related work from human-computer interaction researchers. We conclude with implications for communicative visualization practitioners and researchers, discussing the value of aesthetics and how to achieve it, as well as the need for research on longer-term aesthetic impacts on viewers.

show abstract

“…Gupta et al [13], Rahman et al [41] use a transformer for refining user interface layout aesthetics. Dou et al [10] use CNNs for predicting website aesthetics.…”

Section: Related Workmentioning

confidence: 99%

Interpretable Aesthetic Analysis Model for Intelligent Photography Guidance Systems

Wu¹

2021

Preprint

View full text Add to dashboard Cite

An aesthetics evaluation model is at the heart of predicting users' aesthetic experience and developing user interfaces with higher quality. However, previous methods on aesthetic evaluation largely ignore the interpretability of the model and are consequently not suitable for many human-computer interaction tasks. We solve this problem by using a hyper-network to learn the overall aesthetic rating as a combination of individual aesthetic attribute scores. We further introduce a specially designed attentional mechanism in attribute score estimators to enable the users to know exactly which parts/elements of visual inputs lead to the estimated score. We demonstrate our idea by designing an intelligent photography guidance system. Computational results and user studies demonstrate the interpretability and effectiveness of our method.CCS Concepts: • Human-centered computing → Human computer interaction (HCI); Interaction design.

show abstract

Webthetics: Quantifying webpage aesthetics with deep learning

Cited by 55 publications

References 26 publications

Susceptibility to social influence strategies and persuasive system design: exploring the relationship

Susceptibility to social influence strategies and persuasive system design: exploring the relationship

Aesthetics for Communicative Visualization: a Brief Review

Interpretable Aesthetic Analysis Model for Intelligent Photography Guidance Systems

Contact Info

Product

Resources

About