The XO Platform’s Knowledge AI lets you use FAQs and unstructured documents to improve intent identification for user queries. This feature includes two modules – Knowledge Graph (Ontology-based and Few-Shot) and Answer from Documents.
The Knowledge Graph (KG) lets you turn static FAQ text into an intelligent, personalized conversational experience. There are two types of knowledge graphs:
- The Ontology-based Knowledge Graph lets you create an ontological structure of key domain terms and associate them with context-specific questions and their alternatives, synonyms, and Machine learning-enabled traits.
- The Few-Shot Knowledge Graph leverages Kore.ai’s Large Language Model (LLM) to simplify knowledge organization and KG training, letting you add FAQs to the root node without building and maintaining an ontology structure.
Answer from Documents (Beta)
Answer from Documents helps answer end-user queries from unstructured PDF documents without the need to extract individual FAQs and train them. This feature leverages LLM and Generative AI models from OpenAI or Azure OpenAI to process uploaded documents and user queries and generate answers from the documents. Learn more.