Research Output

How Was Your Day? evaluating a conversational companion.

  The “How Was Your Day” (HWYD) Companion is an embodied conversational agent that can discuss work-related issues, entering free-form dialogues that lack any clearly defined tasks and goals. The development of this type of Companion technology requires new models of evaluation. Here, we describe a paradigm and methodology for evaluating the main aspects of such functionality in conjunction with overall system behaviour, with respect to three parameters: functional ability (i.e., does it do the ‘right’ thing), content (i.e., does it respond appropriately to the semantic context), and emotional behaviour (i.e., given the emotional input from the user, does it respond in an emotionally appropriate way). We demonstrate the functionality of our evaluation paradigm as a method for both grading current system performance, and targeting areas for particular performance review. We show correlation between, for example, ASR performance and overall system performance (as is expected in systems of this type) but beyond this, we show where individual utterances or responses, which are indicated as positive or negative, show an immediate response from the user, and demonstrate how our combination evaluation approach highlights issues (both positive and negative) in the Companion system’s interaction behaviour

  • Type:

    Article

  • Date:

    30 November 2011

  • Publication Status:

    Published

  • Publisher

    IEEE Computer Society

  • DOI:

    10.1109/T-AFFC.2013.15

  • Library of Congress:

    QA75 Electronic computers. Computer science

  • Dewey Decimal Classification:

    006.3 Artificial intelligence

Citation

Benyon, D., Gambäck, B., Hansen, P., Mival, O. & Webb, N. (2011). How Was Your Day? evaluating a conversational companion. IEEE transactions on affective computing. 4, 299-311. doi:10.1109/T-AFFC.2013.15. ISSN 1949-3045

Authors

Keywords

Affectice computing; companions; evaluation; spoken interfaces;

Available Documents