Probabilistic Online Action Prediction
Brian D. Davison, Haym Hirsh · 1998
People display regularities in almost everything they do. This paper proposes characteristics of an idealized algorithm that would allow a user interface to adapt to an individual's pattern of use. We describe a simple predictive method with these characteristics and show its predictive accuracy on a large dataset of UNIX commands to be better than others that have been considered, while using fewer computational and memory resources. Motivation How predictable are you? Each of us displays patterns of actions throughout whatever we do. Most occur without conscious thought. Some patterns are widespread among large communities, and are taught, as rules, such as reading from left to right, or driving on the correct side of the road. Other patterns are a function of our lifestyle, such as picking up pizza on the way home from work every Friday, or programming the VCR to record our favorite comedy each week. Many are a result of the way interfaces are designed, like the pattern of movemen...