Business Record Type Classifier
Sorts a piece of business text into the kind of record it came from: support ticket, email, chat message, CRM note, order note or internal handover note.
Every name, number and address it was trained on is made up. Nothing comes from a real person, patient, customer or business.
Why it exists
A company preparing its records for an AI data license first needs an inventory: how many tickets, emails, chats and CRM notes it holds, and how they link. When exports are mixed (a mailbox with forwarded tickets, a shared drive with pasted chats), sorting by record type is the first step, and each type has its own de-identification needs.
Use it
from transformers import pipeline
classify = pipeline("text-classification", model="aitrainingdatabroker/business-record-type-classifier", top_k=3)
print(classify("Customer says the order arrived with a cracked lid; replacement approved, ship by Friday."))
How it was trained
Fine-tuned from distilroberta-base for 6 epochs on one RTX 3070. Training data: synthetic-deidentification-pairs (480 rows), all invented. The scripts that built the data, trained the model and scored it are in source/; the full scores are in scores.json.
Test scores
On 120 held-out rows: accuracy 1.0, macro F1 1.0.
| Label | F1 | Count |
|---|---|---|
chat |
1.0 | 24 |
crm_note |
1.0 | 17 |
email |
1.0 | 22 |
internal_note |
1.0 | 17 |
order_note |
1.0 | 22 |
support_ticket |
1.0 | 18 |
Limits
- Trained and tested on invented text from the same generators, so its perfect score says little about real exports. Test it on a labelled sample of your own records.
- Six types only. Work orders, change orders, claims files, code reviews and other records are not covered.
- This model is not legal advice and not a certification. Whether a de-identification method meets HIPAA Safe Harbor or Expert Determination, CCPA and CPRA, GDPR or UK GDPR anonymisation, PIPEDA, Quebec Law 25 or another rule is a question for your counsel and, where a rule requires one, a qualified expert. A person should review its output before any data is shared.
Read more
License and credit
Apache 2.0. Please credit AI Training Data Broker and link to https://aitrainingdatabroker.com/ when you share or build on it.
About
AI Training Data Broker is a broker and technical advisor for companies that sell or license their data to AI labs. It guides the technical side, including export, format, de-identification and compliance, and publishes free checklists on de-identifying data and keeping it within the privacy rules that apply.
- Downloads last month
- 29
Model tree for aitrainingdatabroker/business-record-type-classifier
Base model
distilbert/distilroberta-base