An Approach to Generating Human-Computer Interfaces from Task Models.
Matthew L. Bolton, Samaneh Ebrahimi · National Conference on Artificial Intelligence · 2014
Human-computer interaction (HCI) can often occur in situations unanticipated by designers. As such, usability may not always be maintained or human operators may not be able to achieve task goals. This can result in poor system adoption, decreased productivity, and unsafe operating conditions. Formal methods can be used prove properties about different elements of HCI. Unfortunately, these methods require the creation of formal models of interface designs, a non-standard practice that is prone to modeling error. This work describes an approach (based on L* learning) that is being developed to automatically generate formal human-computer interface designs guaranteed to always adhere to usability properties and support human operator tasks. The paper presents a formulation of this approach and describes plans for its implementation. These are discussed along with avenues for future research.