Skip to content
beginnertext

Intent Classification

Classify user utterances into intents for chatbot and virtual assistant training.

Try it live — no install

Boot the real Potato server in your browser (WebAssembly) and annotate with this exact config. Nothing leaves your machine.

Run live in your browser

Get This Design

View on GitHub(opens in a new tab)

Clone or download from the repository

Quick start:

bash
git clone https://github.com/davidjurgens/potato-showcase.git
cd potato-showcase/templates/text/intent-classification
potato start config.yaml

Configuration File

yaml
# Intent Classification Configuration
# Classify user utterances for chatbot training

task_dir: "."
annotation_task_name: "Intent Classification"

data_files:
  - "data/utterances.json"

item_properties:
  id_key: "id"
  text_key: "utterance"
  text_display_key: "utterance"

user_config:
  allow_all_users: true

annotation_schemes:
  - annotation_type: "radio"
    name: "primary_intent"
    description: "What is the user's primary intent?"
    labels:
      - name: "greeting"
        tooltip: "Hello, hi, good morning, etc."
        key_value: "g"
      - name: "goodbye"
        tooltip: "Bye, see you, take care, etc."
        key_value: "b"
      - name: "thanks"
        tooltip: "Thank you, appreciate it, etc."
        key_value: "t"
      - name: "help"
        tooltip: "Request for assistance or information"
        key_value: "h"
      - name: "complaint"
        tooltip: "Expressing dissatisfaction"
        key_value: "c"
      - name: "purchase"
        tooltip: "Wanting to buy something"
        key_value: "p"
      - name: "cancel"
        tooltip: "Wanting to cancel order/subscription"
        key_value: "x"
      - name: "status_check"
        tooltip: "Checking order/delivery status"
        key_value: "s"
      - name: "account_issue"
        tooltip: "Login, password, account problems"
        key_value: "a"
      - name: "refund"
        tooltip: "Requesting money back"
        key_value: "r"
      - name: "feedback"
        tooltip: "Providing feedback or suggestions"
        key_value: "f"
      - name: "other"
        tooltip: "Doesn't fit other categories"
        key_value: "o"

  - annotation_type: "multiselect"
    name: "secondary_intents"
    description: "Any secondary intents present?"
    labels:
      - name: "expressing_urgency"
      - name: "requesting_human"
      - name: "expressing_frustration"
      - name: "asking_for_clarification"
      - name: "providing_information"
      - name: "confirming"
      - name: "declining"

  - annotation_type: "radio"
    name: "sentiment"
    description: "What is the sentiment of the message?"
    labels:
      - name: "Positive"
        key_value: "+"
      - name: "Neutral"
        key_value: "0"
      - name: "Negative"
        key_value: "-"

  - annotation_type: "likert"
    name: "clarity"
    description: "How clear is the user's intent?"
    size: 5
    min_label: "Very unclear"
    max_label: "Very clear"

  - annotation_type: "text"
    name: "entities"
    description: "List any key entities (product names, dates, etc.)"
    required: false

output: "annotation_output/"

output_annotation_dir: "annotation_output/"
output_annotation_format: "json"

Sample Data

json
[
  {
    "id": "intent_001",
    "utterance": "Hi there! I need help with my order.",
    "context": "User just opened chat"
  },
  {
    "id": "intent_002",
    "utterance": "Where is my package? I ordered it a week ago and it still hasn't arrived!",
    "context": "Returning customer"
  }
]

// ... and 6 more items

Details

Annotation Types

likertmultiselectradiotext

Domain

nlpconversational-ai

Use Cases

intent-detectionchatbot-training

Tags

intentchatbotnluvirtual-assistantclassification

Found an issue or want to improve this design?

Open an Issue(opens in a new tab)

Related Designs

Fact Verification

Verify claims as supported, refuted, or not enough information based on provided evidence.

likertmultiselect

AgentBoard Progress Scoring

Assess multi-turn LLM agent progress by identifying achieved milestones, scoring overall progress, categorizing the agent environment, and noting partial progress observations.

multiselectlikert

Audio Transcription Review

Review and correct automatic speech recognition transcriptions with waveform visualization.

likertmultiselect