Abstract
Strategic investments in Artificial Intelligence may enable companies to gain business advantages. There are challenges in using generic natural language processing (NLP) capabilities with complex products and with content that requires specialized domain-specific terminologies. Darwin Information Typing Architecture (DITA)-generated AI schema can leverage enterprise source code to train bots or any other conversational systems to improve the accuracy levels without any manual intervention. A well-defined AI schema is generated from the DITA source files that contain an ontology framework of Intents, Entities, Dialog nodes, along with child nodes, as a result. The schema can be depicted as a JSON file.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Palliyathu, Vishal; Tiwari, Deepti; Kothari, Anu; Upendra, Vineet; and Gupta, Ritu, "DYNAMIC GENERATION OF AN ONTOLOGY-BASED AI SCHEMA FOR CHATBOTS", Technical Disclosure Commons, (June 29, 2020)
https://www.tdcommons.org/dpubs_series/3368