Simulation-Based Usability Evaluation of Spoken and Multimodal Dialogue Systems [electronic resource] / by Stefan Hillmann.
By: Hillmann, Stefan [author.].
Contributor(s): SpringerLink (Online service).
Material type: BookSeries: T-Labs Series in Telecommunication Services: Publisher: Cham : Springer International Publishing : Imprint: Springer, 2018Edition: 1st ed. 2018.Description: XXX, 240 p. 31 illus. online resource.Content type: text Media type: computer Carrier type: online resourceISBN: 9783319625188.Subject(s): Signal processing | User interfaces (Computer systems) | Human-computer interaction | Signal, Speech and Image Processing | User Interfaces and Human Computer InteractionAdditional physical formats: Printed edition:: No title; Printed edition:: No title; Printed edition:: No titleDDC classification: 621.382 Online resources: Click here to access onlineIntroduction -- State of The Art -- MeMo Platform for Automatic Usability Evaluation -- Logging Framework for the Evaluation of Interactive Systems -- User Behaviour Model for the Evaluation of Interactive Systems -- Evaluation of User Behaviour Simulations -- Conclusions.
This book describes an extension of the user behaviour simulation (UBS) of an existing tool for automatic usability evaluation (AUE). This extension is based upon a user study with a smart home system. It uses technical-sociological methods for the execution of the study and the analysis of the collected data. A comparison of the resulting UBS with former UBSs, as well as the empirical data, shows that the new simulation approach outperforms the former simulation. The improvement affects the prediction of dialogue metrics that are related to dialogue efficiency and dialogue effectiveness. Furthermore, the book describes a parameter-based data model, as well as a related framework. Both are used to uniformly describe multimodal human-computer interactions and to provide such descriptions for usability evaluations. Finally, the book proposes a new two-stage method for the evaluation of UBSs. The method is based on the computation of a distance measures between two dialogue corpora and the pair-wise comparison of distances among several dialogue corpora.
There are no comments for this item.