Customer insights with Qdrant, Python and information extractor

by Jimleuk ✓ · 6,832 views · created 2024-08-05 · Market Research, AI Summarization

What this workflow does

This AI workflow has 24 nodes and runs manually, when you click “Execute workflow” (plus 1 other trigger). Along the way it connects Embeddings OpenAI, OpenAI Chat Model, Google Sheets and Qdrant Vector Store. It also calls qdrant:6333 over HTTP.

How it runs

Trigger: Manually, when you click “Execute workflow” (plus 1 other trigger).

1. Set Variables (Edit Fields (Set)): sets 1 field: companyId
2. Set Variables (Edit Fields (Set)): sets 3 fields: companyId, review_date_from, review_date_to
3. Clear Existing Reviews (HTTP Request): calls qdrant:6333 (POST)
4. Find Reviews (HTTP Request): calls an HTTP API (POST)
5. Get TrustPilot Page (HTTP Request): calls an HTTP API
6. Apply K-means Clustering Algorithm (Code): runs custom Python code
7. Extract Reviews (HTML): parses or builds HTML
8. Clusters To List (Split Out): splits a list into separate items
9. Zip Entries (Edit Fields (Set)): sets 1 field: data
10. Only Clusters With 3+ points (Filter): keeps only the items that match a condition

…plus 8 more steps. See the full node list below.

Services it uses

Embeddings OpenAIOpenAI Chat ModelGoogle SheetsQdrant Vector Store

HTTP calls to: qdrant:6333

Scraping pages by hand? Firecrawl is a simpler option (10% off for new users).

Before you upgrade to n8n 3.0

Nodes used (16)

• Manual Triggern8n-nodes-base.manualTrigger
• Edit Fields (Set)n8n-nodes-base.set
• HTMLn8n-nodes-base.html
• Split Outn8n-nodes-base.splitOut
• Default Data Loader@n8n/n8n-nodes-langchain.documentDefaultDataLoader
• Recursive Character Text Splitter@n8n/n8n-nodes-langchain.textSplitterRecursiveCharacterTextSplitter
• Embeddings OpenAI@n8n/n8n-nodes-langchain.embeddingsOpenAi
• HTTP Requestn8n-nodes-base.httpRequest
• OpenAI Chat Model@n8n/n8n-nodes-langchain.lmChatOpenAi
• Filtern8n-nodes-base.filter
• Google Sheetsn8n-nodes-base.googleSheets
• Execute Sub-workflown8n-nodes-base.executeWorkflow
• Execute Workflow Triggern8n-nodes-base.executeWorkflowTrigger
• Qdrant Vector Store@n8n/n8n-nodes-langchain.vectorStoreQdrant
• Coden8n-nodes-base.code
• Information Extractor@n8n/n8n-nodes-langchain.informationExtractor

From the author

This n8n template is one of a 3-part series exploring use-cases for clustering vector embeddings: Survey Insights Customer Insights Community Insights This template demonstrates the Customer Insights scenario where Trustpilot reviews can be quickly grouped by…

View the original on n8n.io → Check your own copy