The HCI has understandably become user-centric, but if we are to consider human operator and computer device as even components of a human-computer system and seek to maximize its overall efficacy with AI methods, we would need to optimize information flows between the two. In the paper, we would like to call to the discussion on defining and measuring the information complexity of modern two-dimensional graphic user interfaces. By analogy with Kolmogorov complexity (algorithmic entropy) for computability resources, the interface information complexity could allow estimating the amount of human processor resources required for dealing with interaction task. The analysis of the current results allows concluding that interface “processing” time by humans is indeed affected by the interface message “length” parameter, and, presumably, by vocabulary size. We hope the results could aid in laying ground for broader AI methods application for HCI in the coming era of ubiquitous Big Interaction.
CITATION STYLE
Bakaev, M., & Avdeenko, T. (2015). Defining and optimizing user interfaces information complexity for AI methods application in HCI. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9170, pp. 397–405). Springer Verlag. https://doi.org/10.1007/978-3-319-20916-6_37
Mendeley helps you to discover research relevant for your work.