How to Create a Knowledge Base: Basic Setup

Basic Setup

General Information

You can define the name of your knowledge base and add descriptions on the following page:

Retrieved Snippets

The number of retrieved snippets represents the maximum number of data snippets that the AI assistant will reference when answering. The system default is "12", meaning the AI assistant will retrieve 12 of the most relevant snippets for each response.

Therefore, you can increase or decrease the number of retrieved snippets to adjust the amount of information the AI assistant references when answering.

What is a Parser?

A Parser enables the system to "understand" the content in PDF documents, making it searchable, editable, or convertible for use in other formats.

When uploading PDF or Word files, the parser offers three options:

MaiAgent Parser (Default)
MaiAgent Parser (Online)
MaiAgent Parser (OCR beta)

The system currently defaults to using MaiAgent Parser, which is particularly suitable for extracting structured data from complex professional documents for AI system analysis.

To understand the differences between the three parsers, please refer to: Technical Manual - Parser Tools

If you encounter issues with data parsing, you can click the [Reparse] icon to have the parser refresh the data.

Retrieval Model Settings

In the knowledge base settings, you can choose which Embedding model and Reranker model you want to use.

Embedding Model

Embedding is like translating human language into "numerical language" that AI can understand, allowing computers to comprehend the true meaning of text. We call this process "vectorization." Different Embedding models have different characteristics, such as the languages they excel at processing and supported deployment environments. Different model settings in the knowledge base can be used to adjust the vectorization processing effect when documents are uploaded. You can choose the most suitable Embedding model for different scenarios.

You can freely choose from multiple Embedding models:

To understand the differences between Embedding models, please refer to: Technical Manual - Embedding Models

Reranker Model

A Reranker is like a professional judge, reassessing which data best answers customer questions from initial search results. What are the differences between using and not using a Reranker?

When a customer asks: "What tent is suitable for beginners? Budget under $8000"

Without Reranker:

AI might answer:
"We have tents at various price points, products at $8000 include..."
(Might mention advanced models, not specifically addressing beginner needs)

With Reranker:

AI answers:
"For beginners, I specially recommend these tents under $8000..."
(Precisely targeting beginner needs + budget + product recommendations)

When search result reranking is enabled, the AI assistant will reorder the retrieved knowledge base content snippets and respond based on the most relevant documents.

To understand Reranker models, please refer to: Technical Manual - Reranker Models

In summary, using Embedding combined with Reranker enables the AI assistant to understand your provided knowledge and review content importance after retrieving snippets, responding with the most relevant knowledge to the question.

Associating AI Assistants

Associating AI assistants means authorizing which AI assistants can use this knowledge base. If you have two AI assistants:

Product Service AI
Order Service AI

When both need to answer questions related to returns, you can associate both AI assistants at once under the "Return Policy" knowledge base settings:

Select the AI assistants to associate

Click to add AI assistant

After adding, it will appear in the selected AI assistants area. Click "Save" in the bottom right to complete the association.

Once associated, both AI assistants can share the "Return Policy" knowledge base, answering based on the same content. Future maintenance only requires updating one knowledge base to ensure AI assistants use the latest data.