The LIMSI RailTel system: field trial of a telephone service for rail travel information
Speech Communication - Special issue on interactive voice technology for telecommunication applications (IVITA '96)
AutoTutor: A simulation of a human tutor
Cognitive Systems Research
| Hi-index | 0.00 |
In sufficiently limited domains, natural language interaction is possible even in the absence of actual natural language understanding. This is particularly true for goal-directed command and control, where the understanding task can essentially be cast as an N-way classification problem. The notion of data-driven semantic inference was recently introduced as an approach to such tasks which in principle allows for unrestricted command/query formulation. This approach relies on a latent semantic analysis framework, whereby each unconstrained word string is automatically mapped onto the intended action through a semantic classification against the set of supported concepts. The objective of this paper is to compare semantic inference with other like-minded N-way classification methods, such as those based on finite-state grammars or nearest neighbor techniques. All experiments are conducted in the context of a desktop user interface control task involving 113 different actions. Results illustrate some of the benefits of semantic inference, specifically in terms of performance and robustness.