Automated workflow
Workflow Description
An intelligent workflow that integrates OpenAI language models with Milvus vector database to create a semantic search and question-answering system. Automatically processes documents, splits them into meaningful chunks, and retrieves information based on stored embeddings.
How it works
- 1.Trigger the workflow manually or through chat interface
- 2.Load and process documents using the default document loader
- 3.Split text into chunks using recursive character segmentation
- 4.Generate embeddings for each chunk via OpenAI API
- 5.Store embeddings in Milvus vector database
- 6.Retrieve relevant information and generate answers using LLM Chain
Use cases
- Build an intelligent document search engine that understands semantic meaning
- Create an automated customer support system answering questions from knowledge base
- Process large document collections and extract information on demand
Requirements
- Active OpenAI API key with sufficient quota
- Configured Milvus vector database instance
- Documents in supported text formats (PDF, TXT, DOCX)
Service Value
Ideal as a smart automation service combining integrations and AI to produce ready-to-use results.
Apps Used
Details
How to Use
- 1.Click "Download Template"
- 2.Open your n8n dashboard
- 3.Go to Workflows > Import from File
- 4.Select downloaded file and configure credentials
Nodes Used (22)
OpenAI Chat Model
OpenAI
When clicking "Execute Workflow"
Manual Trigger
Fetch Essay List
HTTP Request
Extract essay names
HTML Page
Split out into items
Split Out
Fetch essay texts
HTTP Request
Limit to first 3
Limit
Extract Text Only
HTML Page
Sticky Note3
Sticky Note
Sticky Note5
Sticky Note
When chat message received
Chat Trigger
Sticky Note
Sticky Note
Milvus Vector Store
Vector Store Milvus
Default Data Loader
Document Default Data Loader
Embeddings OpenAI
OpenAI
Recursive Character Text Splitter
Text Splitter Recursive Character Text Splitter
Sticky Note1
Sticky Note
Embeddings OpenAI1
OpenAI
Milvus Vector Store1
Vector Store Milvus
Sticky Note2
Sticky Note
Q&A Chain to Retrieve from Milvus and Answer Question
Chain Retrieval Qa
Milvus Vector Store Retriever
Retriever Vector Store