NLP text classification pipeline template

Components in this data flow diagram

• Tickets, emails & reviews
• Ingest via API or queue
• Clean: strip HTML, signatures, PII masking
• Supported language?
• Machine translation
• Tokenise: Hugging Face tokenizer
• Fine-tuned BERT classifier
• Class probabilities
• Top score above 0.8?
• Auto-route: billing, bug, feature request
• Human review queue
• Corrected labels
• Write category to CRM
• Training dataset
• Periodic fine-tuning
• F1 better on holdout set?
• Promote model version
• Keep current model

How to use this data flow diagram template

1. Click Open in the editor. The data flow diagram opens on a new page in FlowDiagram — nothing is uploaded and there is no sign-up.
2. Rename, add or delete shapes and drag them where you want; connectors stay attached as things move.
3. Drop in vendor logos from the library — 1,160 of them, including every AWS, Azure and Google Cloud service icon — or restyle it with the colour and theme controls.
4. Prefer text? Edit the Mermaid source below and paste it in through ☰ → Import diagram (see the Mermaid editor); it imports as an editable diagram.
5. Export to PNG, SVG or PDF, or press Ctrl+L for a link that carries the whole diagram.

Mermaid source

The whole template is the text below. Paste it into FlowDiagram, keep it in Git next to the code it describes, or hand it to an AI assistant to extend.

flowchart LR
  SRC[Tickets, emails & reviews] --> INGEST[Ingest via API or queue]
  INGEST --> CLEAN[Clean: strip HTML, signatures, PII masking]
  CLEAN --> LANG{Supported language?}
  LANG -->|No| TRANSLATE[Machine translation]
  LANG -->|Yes| TOK[Tokenise: Hugging Face tokenizer]
  TRANSLATE --> TOK
  TOK --> MODEL[Fine-tuned BERT classifier]
  MODEL --> PROBS[Class probabilities]
  PROBS --> CONF{Top score above 0.8?}
  CONF -->|Yes| ROUTE[Auto-route: billing, bug, feature request]
  CONF -->|No| HUMAN[Human review queue]
  HUMAN --> LABELS[(Corrected labels)]
  ROUTE --> CRM[Write category to CRM]
  HUMAN --> CRM
  LABELS --> DATASET[(Training dataset)]
  DATASET --> FINETUNE[Periodic fine-tuning]
  FINETUNE --> EVAL{F1 better on holdout set?}
  EVAL -->|Yes| PROMOTE[Promote model version]
  EVAL -->|No| REJECT[Keep current model]
  PROMOTE --> MODEL

Related: Data pipeline diagram tool · Mermaid editor · All ai & ml pipelines templates · All 521 templates