- From: Dave Raggett <dsr@w3.org>
- Date: Tue, 7 Jan 2020 10:39:51 +0000
- To: paoladimaio10@googlemail.com
- Cc: W3C AIKR CG <public-aikr@w3.org>
FYI - I will soon start development on a web-based demo involving a dialogue between three agents that represent a waiter at a restaurant, and a couple of people arriving for a meal together. This scenario was chosen because the language usage and communication intent are familiar and predictable. This will allow me to focus on the cognitive mechanisms for mapping sequences of words to semantic graphs, along with reasoning about likes, dislikes and intent in the context of situational plans. Such mechanisms include the role of spreading activation for disambiguation. The motivation for this work is to prepare the ground for future work on using natural language for teaching cognitive agents everyday common sense skills. > On 7 Jan 2020, at 02:56, Paola Di Maio <paola.dimaio@gmail.com> wrote: > > I have been designing intelligent conversational agents for different purposes > > if someone is interested in the possibility of putting together one at some point for fun but also to experiment and test our possible outputs, we could start evaluating tools/implementation environments. > > For me the requirement is they need to be as simple as possible and web based, > the ones I tried in the past did not work as well as I hoped > suggestions welcome, > Engati? Botpress? suggestions? > available tools keep on changing so not sure what is the state of the art now > > We can put this down as a pet project, for futures, if we manage to evolve an xml set > to monitor/support our own internal CG governance > > PDM > > > > Dave Raggett <dsr@w3.org> http://www.w3.org/People/Raggett W3C Data Activity Lead & W3C champion for the Web of things
Received on Tuesday, 7 January 2020 10:39:55 UTC