Question & Answers on Enterprise Knowledge Base

Food for Thought

AI allows to query proprietary and private data, both unstructured and structured data, using natural language.

Examples

• IT function of a company can offer a chatbot to answer questions about IT policies based on multiple documents and databases
• An e-commerce company can offer a customer self-service to answer questions like “why has not my order shipped yet” by retriving information from different systems and creating an answer through AI
• Create an informational report by combining information from multiple internal data sources

Key Questions

• What internal enterprise information is more convenient to be queried using natural language?
• Is that information structured (e.g. data records in a database or application) or unstructured (e.g. pdf documents)?
• What business processes would benefit from querying proprietary information in natural language?
• Who (which roles) would benefit from querying proprietary information in natural language?

Implementation

Answering questions on proprietary data in natural language can be achieved with so called “Retrieval Augmented Generation”. Check the following best practices for more details.

Related Best Practices

• Vector RAG vs Graph RAG
• Vector-based RAG (1/2) Embedding
• Vector-based RAG (2/2) Query Pipeline
• Knowledge Graph RAG (1/2) Creation
• Knowledge Graph RAG (2/2) Query Pipeline
• Document Grounding