Command line completion: an illustration of learning and decision making using the imprecise Dirichlet model
Erik Quaeghebeur, Gert de Cooman · Ghent University Academic Bibliography (Ghent University) · 2003
A method of command line completion based on probabilistic models is described. The method supplements the existing deterministic ones. The probabilistic models are developed within the context of imprecise probabilities. An imprecise Dirichlet model is used to represent the assessments about all possible completions and to allow for learning by observing the commands typed previously. Due to the use of imprecise probabilities a partial (instead of a linear) ordering of the possible completion actions will be constructed during decision making. Markov models can additionally be incorporated to take recurring sequences of commands into account.