Saturday, October 14, 2023

Intro to AI Workshop: Excel with AI Before the AI Apocalypse

 An ontology is your core concepts and their relationships to the world, according to Noz Urbina. They are areas of concepts with specific, meaningful (semantic) relationships. When populated from data, they form a knowledge graph.

These relationships are not easy to set up in traditional content management systems. You can use ontology management software.

If you have an ontology, you can support auto-tagging. Tags can be mined, to discover what you talk about frequently. If you have a large amount of content, any amount of automation is huge.

You can have all the assets your customers could want, but if they cannot be found, they are useless. 

Tags represent concepts. But you don't have to tag every concept because in your ontology, you define the relationships to your tags. Those relationships help the AI understand the content it is ingesting. 

Generative design is like working with an all-powerful, really painful stupid genie. Can be both magical and over-literal. You have to be very careful with how you word your questions. 

AI will not model your data for you. AI is getting better at structuring unstructured content, but it still can't be trusted. Applying a structure is not as hard as defining a structure. Defining one structure is easier than defining a model. 

AI enables us to be programmers without being coders. 

A journey is questions over time. It's no more complex than that. (If you want, you can multiply the questions factor by emotions.) And then to get even a bit more fancy, Journey = Context + Questions x Emotions / Time. 

Measuring satisfaction of customer journeys is better than measuring satisfaction of specific interactions.


No comments:

Post a Comment

Closing Panel Discussion: The Future of Content

5 year ago, chatbots, and then nothing. And then metaverse, and then nothing. Now, chatbots. So how to prepare for the next 6 months.  Self-...