⚙️

Confidence Score

ActiveCoreMUSTFREE

What it does

Every HS code classification includes a multi-dimensional confidence score from 0 to 1. The score reflects how precisely the product maps to the assigned HS code based on input completeness, rule match strength, and ambiguity level. A score above 0.9 indicates high confidence; below 0.7 suggests the product description may be too vague.

Required Fields for 100% Accuracy

1 required / 4 total

Field	Type	Required	Description	Example
`productName`	string	!	Product name — more detail yields higher confidenceInclude material + form + function for best scores.	`Stainless steel insulated water bottle`
`category`	string	opt	Product categoryBoosts score by narrowing Section scope.	`Drinkware`
`material`	string	opt	Primary material (WCO group)Directly maps to HS Sections XI–XV. Critical for scores above 0.9.	`stainless steel`
`processing`	string	opt	Manufacturing methodDifferentiates subheadings — e.g., cast vs forged steel.	`pressed`

Tips for Best Accuracy

✓More input fields = higher confidence score
✓productName alone typically yields 0.6–0.8 confidence
✓Adding material + processing pushes most items above 0.9
✓Scores below 0.7 usually mean the product name is too generic

Common Mistakes

✗Expecting 1.0 confidence with only productName — provide at least 3 fields
✗Ignoring the alternatives array — the second-best option may be more accurate for your specific product variant

How to use it

Classify a product

Send a classification request via POST /api/v1/classify with as much detail as possible.

Read the confidence field

The response includes a confidence score between 0 and 1.

Check alternatives

The alternatives array shows other possible HS codes with their own confidence scores.

Improve low scores

If confidence is low, add more fields: material, processing method, intended use, or composition.

API Reference

POST/api/v1/classify

Request

{
  "productName": "Stainless steel water bottle",
  "category": "Drinkware"
}

Response

{
  "hsCode": "7323.93",
  "description": "Table, kitchen or other household articles, of stainless steel",
  "confidence": 0.92,
  "alternatives": [
    {
      "hsCode": "7310.29",
      "confidence": 0.71
    },
    {
      "hsCode": "7612.90",
      "confidence": 0.55
    }
  ]
}

cURL

curl -X POST https://potal.app/api/v1/classify \
  -H "X-API-Key: pk_live_your_key" \
  -H "Content-Type: application/json" \
  -d '{"productName":"Stainless steel water bottle","category":"Drinkware"}'

Related Features

⚙️HS Code Classification

Classify products into 5,000+ HS codes using the v3.3 GRI pipeline

⚙️Audit Trail

Complete classification history with decision path and rules applied

⚙️HS Code Validation

9-field input validation ensuring accurate classification inputs

⚙️Batch Classification

Classify hundreds of products at once via CSV or JSON batch API

⚙️Origin Detection

Detect country of origin from product description and brand analysis

Having issues?

🐛 Report a Bug ❓ Ask a Question 💡 Suggest Improvement