Model Based Estimation for Multi-Modal User Interface Component Selection

Multi-modal human computer interfaces are becoming increasingly widespread, building on more capable and affordable devices along with advances in helper applications utilising advanced pattern recognition. These interfaces promise to improve human-computer interaction, not only for fully-abled people, but also for persons with disabilities. It is currently unclear how to map available multi-modal components to a specific user profile in a systematic fashion, especially when the abilities, perceptual preferences and literacy level of the user should be taken into account. This paper presents one approach to develop a cost-based model which can be used to derive appropriate mappings for specific user profiles. The model is explained through a number of small examples, where after the usage and benefits of the model are illustrated using a variety of different profiles. It is shown that the model is effective in identifying important multi-modal components for various user profiles.