balzanilo/dialogs-mtl-dataset

---
language:
- en
license: mit
size_categories:
- 1K<n<10K
task_categories:
- text-classification
pretty_name: Emotions/Triggers in multi-speaker dialogues
dataset_info:
features:
- name: speakers_utterances
sequence: string
- name: emotions
sequence: int64
- name: triggers
sequence: float64
splits:
- name: train
num_bytes: 1956813
num_examples: 3187
- name: test
num_bytes: 270767
num_examples: 419
- name: eval
num_bytes: 239586
num_examples: 394
download_size: 644361
dataset_size: 2467166
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: eval
path: data/eval-*
tags:
- emotion_classification
- trigger_detection
---
# Conversation Emotion and Trigger Dataset
The Conversation Emotion and Trigger Dataset is a collection of conversational exchanges annotated with emotions and triggers. Each instance in the dataset consists of utterances, speaker utterances, corresponding emotions, and triggers.

## Columns
* Utterances (List of Strings): This column contains the textual content of each utterance within the conversation.
* Speaker Utterances (List of Strings): This column contains the speaker's name followed by the corresponding utterance.
* Emotions (List of Integers): Each integer in this column represents the emotion associated with the corresponding utterance. Emotions are encoded as follows:
- 0: Neutral
- 1: Surprise
- 2: Fear
- 3: Sadness
- 4: Joy
- 5: Disgust
- 6: Anger
* Triggers (List of Integers): This column indicates whether the corresponding utterance triggers a modification in emotion. A value of 1 denotes a trigger, while 0 denotes no trigger.

### Example Instance
#### Utterances
```py
[
"Hey.",
"Hey!",
"So how was Joan?",
"I broke up with her.",
"Don't tell me, because of the big nostril thing?",
"They were huge. When she sneezed, bats flew out of them.",
"Come on, they were not that huge.",
"I'm tellin' you, she leaned back; I could see her brain.",
"How many perfectly fine women are you gonna reject over the most superficial insignificant things?"
]
```

#### Speaker Utterances
```py
[
"Chandler: Hey.",
"All: Hey!",
"Monica: So how was Joan?",
"Chandler: I broke up with her.",
"Ross: Don't tell me, because of the big nostril thing?",
"Chandler: They were huge. When she sneezed, bats flew out of them.",
"Rachel: Come on, they were not that huge.",
"Chandler: I'm tellin' you, she leaned back; I could see her brain.",
"Monica: How many perfectly fine women are you gonna reject over the most superficial insignificant things?"
]
```

#### Emotions
```py
[0, 4, 0, 0, 1, 5, 0, 5, 1]
```

#### Triggers
```py
[0, 0, 0, 0, 0, 0, 0, 1, 0]
```

原始信息汇总

对话情感与触发数据集

数据集概述

对话情感与触发数据集是一个包含对话交流并标注情感和触发因素的集合。每个实例包含话语、说话者话语、相应的情感和触发因素。

数据集信息

• 语言: 英语
• 许可证: MIT
• 大小分类: 1K<n<10K
• 任务分类: 文本分类
• 美观名称: Emotions/Triggers in multi-speaker dialogues

特征

• speakers_utterances: 序列类型为字符串
• emotions: 序列类型为int64
• triggers: 序列类型为float64

分割

• train: 字节数为1956813，实例数为3187
• test: 字节数为270767，实例数为419
• eval: 字节数为239586，实例数为394

大小

• 下载大小: 644361字节
• 数据集大小: 2467166字节

配置

• config_name: default
  • data_files:
    • train: data/train-*
    • test: data/test-*
    • eval: data/eval-*

标签

• emotion_classification
• trigger_detection

列信息

• Utterances: 包含对话中每个话语的文本内容。
• Speaker Utterances: 包含说话者的名字和相应的话语。
• Emotions: 每个整数代表相应话语的情感，编码如下：
  • 0: 中性
  • 1: 惊讶
  • 2: 恐惧
  • 3: 悲伤
  • 4: 喜悦
  • 5: 厌恶
  • 6: 愤怒
• Triggers: 指示相应话语是否触发情感变化。1表示触发，0表示无触发。

示例实例

Utterances

python [ "Hey.", "Hey!", "So how was Joan?", "I broke up with her.", "Dont tell me, because of the big nostril thing?", "They were huge. When she sneezed, bats flew out of them.", "Come on, they were not that huge.", "Im tellin you, she leaned back; I could see her brain.", "How many perfectly fine women are you gonna reject over the most superficial insignificant things?" ]

Speaker Utterances

python [ "Chandler: Hey.", "All: Hey!", "Monica: So how was Joan?", "Chandler: I broke up with her.", "Ross: Dont tell me, because of the big nostril thing?", "Chandler: They were huge. When she sneezed, bats flew out of them.", "Rachel: Come on, they were not that huge.", "Chandler: Im tellin you, she leaned back; I could see her brain.", "Monica: How many perfectly fine women are you gonna reject over the most superficial insignificant things?" ]

Emotions

python [0, 4, 0, 0, 1, 5, 0, 5, 1]

Triggers

python [0, 0, 0, 0, 0, 0, 0, 1, 0]