Text classification

Natural Language Processing (NLP) is an area of research and application that explores how to use computers to “understand” and manipulate natural language, such as text or speech. Most NLP techniques rely on machine learning to derive meaning from human languages. One of NLP’s methodologies for processing natural language is text classification, a method that leverages deep learning to categorize sequences of unstructured text.

Here are some ways to use text classification:

  • classify user sentiment in a review
  • flag inappropriate content
  • optimize marketing efforts
  • etc

We advise that you invest enough time pre-processing your data and configuring your ontology to avoid flaws or irregularities in your labeled dataset.

The custom text classification labeling interface is only available in the legacy image editor.

Configuring text classification

  1. Create a JSON file and upload it to Labelbox
// Example dataset
"externalId": "001",
"data": "Customer-\nI love this tool and the customer service.\n\nOperator-\nThanks so much. We are here to help."
"externalId": "002",
"data": "Customer-\nI am having trouble with sending a payment.\n\nOperator-\nI am sorry to hear that. Can you share if you are using a chrome or firefox browser?"
  1. Create a project
  2. Attach your JSON file as your dataset
  3. Install a custom template for your label editor by entering https://classification.labelbox.com/ as the URL to the label editor.
  4. Configure your ontology
// Example project ontology
"tools": [],
"classifications": [
"name": "appearance",
"instructions": "Is this customer happy with the service?",
"type": "radio",
"options": [
"value": "yes",
"label": "Yes"
"value": "no",
"label": "No"

Was this page helpful?

Named entity recognition