Context Prediction is The Key For Conversational Chatbots

Andy McDonald
Follow me

Andy McDonald

Trained as a social psychologist, Andy McDonald has been designing and writing documentation for the oil industry since 1998, and is now Product Manager for Tech Advantage in the Paris area.
Having seen methods, norms and formats come and go, his basic training leads him naturally to concentrate on the people involved in the processes and the end user requirements.
Andy McDonald
Follow me

Let´s dive deeper in the topic of context and contextualization:

  1. What is Contextualization all About & Why is it so Important in Information 4.0?
  2. Context is Key for Conversational Chatbots

At the Information Energy Day at tcworld in Suttgart, the level of exchanges between attendees and speakers demonstrates that chatbots are indeed the technology that a lot of companies are trying to embrace. It also confirmed that professionals from the techcomm world have a key role to play alongside the technologists putting them together.

Contextualize The Bot

One question dealt with, that I would summarize as “Can we have multiple purpose-oriented bots or is the bot multipurpose?”, really shows the high level of thinking going on around bot deployment. Bots should probably be targeted for an audience and a purpose such as onboarding, support, sales, service. Then the next question is going to be about the general welcome bot acting as the foyer for highly quality reception and orientation.

In this respect we can imagine that we need to pass information from one bot to another. This information is a context matrix. For the moment, context is collected in the bot. Tomorrow it may come from contextualizing engines.

Context Detection And Prediction

Can we use this matrix to write molecules for contextualization in the future?

context engine stephan sigg intelligent information

The nature of the input data, the quality output, and the flow of processing operations that are used to make a prediction, are correlated.

Bot Content Is A Narrative

carlos perez cognitive stack

Carlos E. Perez about ‘Cognitive Stack for Artifical Intelligence’ at tcworld 17 conference.

Keeping people in the bot is having and engaging narrative, a conversation. As we move forward this conversation may be two-way, but in reality it may not be any time soon. We learned a bit more about the reality of AI & deep learning from Carlos Perez from Intuition Engine. Having a free form conversation between an AI driven bot and humans is quite simple a long time away still. It will require cognition in AI. Deep learning can be put into play more easily.

 

Context Is Everywhere

At the Information Energy day and in other exchanges at tcworld, the notion of context kept turning up. As an industry it is essential that we move forward in this area. It is one of the key elements that will help us transform dislocated conversations in bots (however useful they are) into the emergence of what Joe Gartner calls the intelligent personal assistant.

So, at Information Energy 2018, we will organize an open workshop on March 2nd on “How do we want to contextualize?” with Stephan Sigg and Carlos Perez and Steffen Frederiksen. The ambition is helping us define the next proof of concept.

stephan-sigg-keynote-ai-context-chatbots

Real-World Examples Of Implementation Challenge Us

Some of the implementation examples really challenge some of the thinking in the consortium and we need to keep this going. If you have examples come show them. Participate and send us your ideas and come help us move to the next step.

If you are in technology that is going to change the future come tell us what you are doing.

 Submit your paper now 

[Total: 2    Average: 4.5/5]

Spread the word. Share this post!

Leave A Reply

Your email address will not be published. Required fields are marked *