標註方案

Name: Potato
Author: Potato

定義標註者將如何標註您的資料。

標註方案定義了標註者的標註任務。Potato 支援 30 多種標註類型，可以組合建立複雜的標註任務。

標註方案剖析 — 四個欄位定義每個控制元件；堆疊方案即可對每個條目提多個問題標註方案剖析

基本結構

每個方案在 annotation_schemes 陣列中定義：

yaml

annotation_schemes:
  - annotation_type: radio
    name: sentiment
    description: "What is the sentiment?"
    labels:
      - Positive
      - Negative
      - Neutral

必填欄位

欄位	描述
`annotation_type`	標註類型（radio、multiselect、likert、span、text、number、slider、multirate）
`name`	內部識別符號（無空格，用於輸出）
`description`	顯示給標註者的說明

支援的標註類型

1. 單選（Radio）

從列表中選擇一個選項：

yaml

- annotation_type: radio
  name: sentiment
  description: "What is the sentiment of this text?"
  labels:
    - Positive
    - Negative
    - Neutral
 
  # Optional features
  sequential_key_binding:
    Positive: "1"
    Negative: "2"
    Neutral: "3"
 
  # Or use sequential binding (1, 2, 3... automatically)
  sequential_key_binding: true
 
  # Horizontal layout instead of vertical
  horizontal: true

2. 李克特量表（Likert Scale）

帶有端點標籤的評分量表：

yaml

- annotation_type: likert
  name: agreement
  description: "How much do you agree with this statement?"
  size: 5  # Number of scale points
  min_label: "Strongly Disagree"
  max_label: "Strongly Agree"
 
  # Optional mid-point label
 
  # Show numeric values

3. 多選（Multiselect）

從列表中選擇多個選項：

yaml

- annotation_type: multiselect
  name: topics
  description: "Select all relevant topics"
  labels:
    - Politics
    - Technology
    - Sports
    - Entertainment
    - Science
 
  # Selection constraints
 
  # Allow free text response
  has_free_response: true

4. 片段標註（Span Annotation）

高亮並標記文本片段：

yaml

- annotation_type: span
  name: entities
  description: "Highlight named entities in the text"
  labels:
    - PERSON
    - ORGANIZATION
    - LOCATION
    - DATE
 
  # Visual customization
 
  # Allow overlapping spans
  allow_discontinuous: false
 
  # Keyboard shortcuts for labels
  sequential_key_binding: true

5. 滑塊（Slider）

連續數值範圍：

yaml

- annotation_type: slider
  name: confidence
  description: "How confident are you in your answer?"
  min_value: 0
  max_value: 100
  step: 1
 
  # Endpoint labels
 
  # Show current value

6. 文本輸入（Text Input）

自由文本回復：

yaml

- annotation_type: text
  name: explanation
  description: "Explain your reasoning"
 
  # Multi-line input
  rows: 4
 
  # Character limits
 
  # Placeholder text
  placeholder: "Enter your explanation here..."
 
  # Disable paste (for transcription tasks)

7. 數字輸入（Number Input）

帶約束的數字輸入：

yaml

- annotation_type: number
  name: count
  description: "How many entities are mentioned?"
  min: 0
  max: 100
  step: 1

8. 多項評分（Multirate）

在同一量表上對多個項目進行評分：

yaml

- annotation_type: multirate
  name: quality_aspects
  description: "Rate each aspect of the response"
 
  # Randomize item order
 
  # Layout options

通用選項

鍵盤快捷鍵

通過鍵盤繫結加速標註：

yaml

# Manual shortcuts
keyboard_shortcuts:
  Positive: "1"
  Negative: "2"
  Neutral: "3"
 
# Or automatic sequential binding
sequential_key_binding: true  # Assigns 1, 2, 3...

工具提示

為標籤提供懸停提示：

yaml

tooltips:
  Positive: "Expresses happiness, approval, or satisfaction"
  Negative: "Expresses sadness, anger, or disappointment"
  Neutral: "No clear emotional content"

標籤顏色

自定義顏色以進行視覺區分：

yaml

label_colors:
  PERSON: "#3b82f6"
  LOCATION: "#10b981"
  ORGANIZATION: "#f59e0b"

必填欄位

在提交前使方案成為必填項：

yaml

- annotation_type: radio
  name: sentiment
  label_requirement:
    required: true
  labels:
    - Positive
    - Negative

多方案組合

每個實例組合多種標註類型：

yaml

annotation_schemes:
  # Primary classification
  - annotation_type: radio
    name: sentiment
    description: "Overall sentiment"
    labels:
      - Positive
      - Negative
      - Neutral
    label_requirement:
      required: true
    sequential_key_binding: true
 
  # Confidence rating
  - annotation_type: likert
    name: confidence
    description: "How confident are you?"
    size: 5
    min_label: "Guessing"
    max_label: "Certain"
 
  # Topic tags
  - annotation_type: multiselect
    name: topics
    description: "Select all relevant topics"
    labels:
      - Politics
      - Technology
      - Sports
      - Entertainment
    has_free_response: true
 
  # Notes
  - annotation_type: text
    name: notes
    description: "Any additional observations?"
    rows: 4
    label_requirement:
      required: false

高階功能

成對比較

比較兩個項目：

yaml

- annotation_type: pairwise
  name: preference
  description: "Which response is better?"
  labels:
    - label: "Response A"
      value: "A"
    - label: "Response B"
      value: "B"
    - label: "Equal"
      value: "tie"
 
  # Allow tie selection
  allow_tie: true

最佳-最差縮放

通過選擇最佳和最差來排列項目：

yaml

- annotation_type: bws
  name: ranking
  description: "Select the best and worst items"
  # Items come from the data file

下拉選擇

節省空間的單選：

yaml

- annotation_type: select
  name: category
  description: "Select a category"
  labels:
    - Category A
    - Category B
    - Category C
    - Category D
    - Category E
 
  # Default selection

資料格式參考

輸入

標註方案與您的資料格式配合使用：

json

{
  "id": "doc_1",
  "text": "This is the text to annotate."
}

輸出

標註以方案名稱為鍵儲存：

json

{
  "id": "doc_1",
  "annotations": {
    "sentiment": "Positive",
    "confidence": 4,
    "topics": ["Technology", "Science"],
    "entities": [
      {"start": 0, "end": 4, "label": "ORGANIZATION", "text": "This"}
    ],
    "notes": "Clear positive sentiment about technology."
  }
}

最佳實踐

1. 清晰的標籤

使用明確、有區分度的標籤：

yaml

# Good
labels:
  - Strongly Positive
  - Somewhat Positive
  - Neutral
  - Somewhat Negative
  - Strongly Negative
 
# Avoid
labels:
  - Good
  - OK
  - Fine
  - Acceptable

2. 有用的工具提示

為有細微差別的標籤新增工具提示：

yaml

tooltips:
  Sarcasm: "The text says the opposite of what it means"
  Irony: "A mismatch between expectation and reality"

3. 鍵盤快捷鍵

為大批次任務啟用快捷鍵：

yaml

sequential_key_binding: true

4. 邏輯順序

始終如一地排列標籤：

最常見的放在首位
按字母順序排列
按強度排列（從低到高）

5. 限制選項數量

過多的選擇會減慢標註速度：

單選：2-7 個選項
多選：5-15 個選項
李克特量表：5-7 個等級

6. 先行測試

在部署之前自己標註幾個示例，以發現：

含糊的標籤
遺漏的類別
不清晰的說明