Create a Knowledge Base with Knowledge Pipelines

A knowledge pipeline is a solution for building document processing workflows. You can visually combine and configure different nodes, orchestrating them like a workflow, and choose various tools to optimize the data processing.

It mainly consists of four stages, each made up of different nodes and tools, forming a complete data processing chain:

Data Source → Document Conversion → Text Chunking → Knowledge Base Storage

Each step serves a specific purpose: collecting content from various sources, converting it into processable text, optimizing it for search, and storing it in a format that enables fast and accurate retrieval.

Additionally, XpertAI provides knowledge pipeline templates for different use cases, helping to improve the accuracy of data indexing and retrieval results. This section will help you understand how to create knowledge pipelines, their processes, and the corresponding nodes, so you can quickly build and optimize your own knowledge base.

Step-by-Step Guide

• Step 1: Create a Knowledge Pipeline
  Start from a built-in template, a blank knowledge pipeline, or import an existing pipeline.

• Step 2: Orchestrate the Knowledge Pipeline
  Learn how the knowledge pipeline works, orchestrate different nodes, and build the data processing workflow you need.

• Step 3: Publish the Knowledge Pipeline
  After testing the configuration, publish the pipeline to prepare for document processing.

• Step 4: Upload Files
  Add documents, which will be processed to build a searchable knowledge base.

• Step 5: Manage and Use the Knowledge Base
  Maintain documents, test retrieval results, modify settings, and more.