eXvisory dev is a web-based development framework for building deep logic product support chatbots able to troubleshoot complex problems. Its main component is a visual editor based around a fault containment hierarchy.
eXvisory dev web editor (fault diagnosis)
The concepts you need to understand to use eXvisory dev are deliberately simple, to enable you to build deep logic networks without being overwhelmed by accumulating complexity.
You create and configure logic rules from a limited palette of logic rule types (faults, fault groups, eliminators, query tests and synthetic tests) and fit them together into a visual fault containment hierarchy that models how experts troubleshoot product support issues. This creates a deep logic network from which eXvisory dev automatically generates a chatbot. Inference is the route eXvisory takes through the fault containment hierarchy (and its underlying deep logic network) so that the chatbot asks the right questions in the right order with relevant help and explanations.
Most product documentation starts with a simple "Hello World" example. But eXvisory dev is about enabling you to build and maintain complex deep logic networks, so we thought we'd take a different approach and start with an easy to understand but more complex example.
Browse to https://dev.exvisory.ai/apps/sample-mobile to open eXvisory dev, loaded with a read-only snapshot of our pilot mobile phone and tablet troubleshooting deep logic network.
For access to https://dev.exvisory.ai/apps/sample-mobile you have to register and be accepted into our developer program.
As you read through the documentation keep the sample-mobile deep logic network open in another browser window or tab so you can experiment with eXvisory dev while learning by example about each of its concepts.
- Inference - learn how the fault containment hierarchy works and the order in which fault nodes are evaluated.
These introductory pages are designed to give you a good idea of how eXvisory dev works and how you can use it to develop your own chatbot. See the Reference pages for the gory details.