ArmanEmo: a Persian dataset for text-based emotion detection.

The rapid growth of textual data on social media platforms has increased attention to emotion detection (ED) from text. Businesses and online service providers leverage ED techniques to analyze customer or user sentiments toward their products and services, enabling more informed and strategic decis...

Descripción completa

Detalles Bibliográficos
Publicado en:Language Resources & Evaluation Vol. 59; no. 3; pp. 2565 - 2588
Autores principales: Mirzaee, Hossein, Peymanfard, Javad, Habibzadeh Moshtaghin, Hamid, Zeinali, Hossein
Formato: Artículo
Publicado: Springer Nature Sep2025
Materias:
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=hlh&AN=186909076&site=ehost-live
header:
  @attributes:
    shortDbName: hlh
    uiTerm: 186909076
    longDbName: Humanities International Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    jinfo:
      jid:
        1574020X
        179V
      jtl: Language Resources & Evaluation
      issn: 1574020X
      maglogo: N
    pubinfo:
      dt: Sep2025
      vid: 59
      iid: 3
      pid: 237
      pub: Springer Nature
    artinfo:
      ui:
        186909076
        10.1007/s10579-025-09817-4
      ppf: 2565
      ppct: 23
      formats:
        fmt:
          – @attributes:
              type: T
          – @attributes:
              type: P
              size: 1.4MB
      tig:
        atl: ArmanEmo: a Persian dataset for text-based emotion detection.
      aug:
        au:
          Mirzaee, Hossein
          Peymanfard, Javad
          Habibzadeh Moshtaghin, Hamid
          Zeinali, Hossein
        affil:
          https://ror.org/04gzbav43 Department of Chemical Engineering, Amirkabir University of Technology, Tehran, Iran
          https://ror.org/01jw2p796 School of Computer Engineering, Iran University of Science and Technology, Tehran, Iran
          https://ror.org/02cc4gc68 Faculty of Management and Accounting, Allameh Tabataba'i University, Tehran, Iran
          https://ror.org/04gzbav43 Department of Computer Engineering, Amirkabir University of Technology, Tehran, Iran
      su:
        Emotion recognition
        Emotions
        Transformer models
        Sentiment analysis
        Transfer of training
        Content analysis
        Social media
      sug:
        subj:
          Emotion recognition
          Emotions
          Transformer models
          Sentiment analysis
          Transfer of training
          Content analysis
          Social media
      keyword:
        BERT model
        Ekman's model
        Emotion detection
        Persian dataset
        Psychology and Cognitive Sciences Psychology
      ab: The rapid growth of textual data on social media platforms has increased attention to emotion detection (ED) from text. Businesses and online service providers leverage ED techniques to analyze customer or user sentiments toward their products and services, enabling more informed and strategic decision-making. In this study, we introduce ArmanEmo, a human-labeled emotion dataset of more than 7000 Persian sentences labeled for seven categories. The dataset has been collected from different resources, including Twitter, Instagram, and Digikala comments. The labels are based on Ekman's six basic emotions-Anger, Fear, Happiness, Hatred, Sadness, and Wonder-augmented by an additional category, "Other", to account for emotions outside Ekman's model. In addition to the dataset, we have provided several baseline models for emotion classification, focusing on the state-of-the-art transformer-based language models. Our best model achieves a macro-averaged F1 score of 75.39% across our test dataset. Moreover, we also conduct transfer learning experiments to evaluate how well models trained on our proposed dataset perform on unseen samples from another Persian emotion dataset, EmoPars. Results of these experiments suggest that the model trained on our training set has a superior performance to the model trained on the training set of EmoPars, even when the models are evaluated on the test set of EmoPars. ArmanEmo is publicly available for non-commercial use at https://github.com/Arman-Rayan-Sharif/arman-text-emotion.
      pubtype: Academic Journal
      doctype: Article
      src: R
    language: English
    refInfo:
    copyright:
      @attributes:
        flag: Y
      custom: Language Resources & Evaluation is a copyright of Springer, 2025. All Rights Reserved.
      item: Language Resources & Evaluation
      holder: Springer Nature
      dt:
        @attributes:
          year: 2025
    holdings:
      @attributes:
        islocal: N