iberbench/tass2020_emotion_analysis

Dataset Card for TASS 2020 Emotion Analysis

Dataset Summary

The TASS 2020 dataset for Task 2 focuses on emotion analysis in Spanish text, specifically identifying the presence and intensity of emotions such as anger, joy, sadness, fear, and surprise. The data originates from Spanish-language social media, providing a challenging and rich resource for emotion detection.

Supported Tasks and Leaderboards

Emotion detection, consists of classifying text into predefined emotion categories such as anger, joy, sadness, fear, and surprise. Success on this task is typically measured by achieving a high F1-score (https://huggingface.co./metrics/f1)."

Languages

The language in the dataset is Spanish as used in social media platforms by Spanish-speaking users. The text reflects informal language with common usage of abbreviations, slang, and emojis typical of online communication. The BCP-47 code for Spanish is es.

Dataset Structure:

Data Instances:

For each instance, there is a string representing the text (a social media post) and a list of emotions detected in the text

{'text':'El Atlético resignado a perder HASHTAG 😔 https://t.co/AiaaGdxmQO'
 'label':'sadness'}

Data Fields

• text: The original social media text.
• label: The specific emotion detected
• language: language of the text.

Data splits:

Dataset Split	Number of Instances in Split
Train	5,890
Test	857

Dataset Creation:

Curation rationale

Task 2 of TASS 2020 was designed to provide a benchmark for emotion analysis in Spanish, addressing the need for understanding nuanced emotional expression in social media text.

Source data

Initial Data Collection and Normalization

The data was collected from Spanish social media platforms and processed to anonymize personal details and normalize text for analysis.

Who are the source language producers?

Spanish-speaking users of social media. No demographic information about the producers is available.

Annotations

Annotation process

Emotion labels and intensities were annotated by Spanish-speaking experts. Each instance was reviewed by multiple annotators for quality control.

Annotators

Annotations were conducted by experts in linguistics and natural language processing.

Personal and sensitive information

The dataset is anonymized and contains no identifiable personal information.

Considerations for Using the Data

Under no circumstances we claim this dataset as ours, all merits are destined towards the emotion analysis TASS 2020 workshop authors.

• social impact of dataset: This dataset enables the development of models capable of understanding emotions expressed in Spanish, which can improve applications in mental health, customer service, and social media monitoring.
• discussion of biases: The dataset may reflect biases inherent in social media, such as overrepresentation of certain demographics or topics.
• other known limitations: The dataset is derived from short social media text and may not generalize to other domains or more formal styles of Spanish.

Addtional Information

Dataset Curators

The dataset was curated by the TASS 2020 organizers and collaborators from SEPLN.

licensing_information:

Available under License Information Needed.

Citation Information

      @inproceedings{tass2020_emotion_analysis,
          title = "{TASS} 2020 Task 2: Emotion Analysis in Spanish Social Media Text",
          author = "Villena-Román, Julio  and Correa, Elizabeth",
          booktitle = "Proceedings of the TASS 2020 Workshop",
          year = "2020",
          address = "Alicante, Spain",
          publisher = "SEPLN",
          url = "http://tass.sepln.org/2020/"
      }