Re: Virtual Agent framework?

One question that comes to mind is whether there is a relatively static 
domain model,  separate from the more dynamic data model (in a language 
like SCXML).  The data model is used to keep track of the state of the 
current interaction (number of items in the shopping cart, etc.) while 
the domain/task model would represent the basic features of all 
interactions.

Another question is whether we think of the virtual agents as having 
scripts that they use to interact with users.  (Contact centers often 
assign scripts to actual human agents to guide their interactions with 
callers.)  Scripts would be like interaction templates.   Again thinking 
in terms of a language like SCXML, scripts could be state machines.  If 
the virtual agent has multiple scripts to choose from, we can either 
represent that as one big state machine (with branching logic at the 
top) or we can break it down so that one component would  select the 
script while another would execute it.

- Jim

On 8/24/2016 9:02 AM, Deborah Dahl wrote:
>
> I’d like to talk about putting together a framework for virtual 
> assistants along the lines of  the Speech Interface Framework 
> (https://www.w3.org/TR/voice-intro/ see Fig. 1) or the Multimodal 
> Framework (https://www.w3.org/TR/mmi-framework/ see Figs. 2 and 3). It 
> would list the components of a virtual agent/community of virtual 
> agents and point out where existing and/or future standards can be 
> used. I’ll try to draft a diagram that we can talk about. If you have 
> any thoughts about this, please post to the list.
>
>
>
> ------------------------------------------------------------------------
> Avast logo <https://www.avast.com/antivirus> 	
>
> This email has been checked for viruses by Avast antivirus software.
> www.avast.com <https://www.avast.com/antivirus>
>
>