Skip to content

This dataset maps multiple positive emotional elements to musical features from classical and New Age music. Validated by experts, it supports emotion recognition, social-emotional learning, and therapeutic music applications, especially for special needs populations.

License

Notifications You must be signed in to change notification settings

Xeowa/A-Social-Emotional-Music-DataSet

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 

Repository files navigation

SEM Music Dataset: A Dataset for Social-Emotional Music Classification

Kaggle Dataset

Get the Data

Abstract

The Social-Emotional Music (SEM) Dataset represents a novel contribution to music information retrieval research, specifically addressing the intersection of music psychology and social-emotional learning (SEL). This curated collection provides music tracks systematically categorized according to three fundamental emotional competencies: Outlook, Problem-Solving, and Empathetic Perspective-Taking.

Unlike traditional music datasets that focus primarily on genre classification or basic emotional valence, the SEM Dataset bridges the gap between low-level acoustic features and high-level psychological constructs. The dataset was developed through interdisciplinary collaboration between musicologists, psychologists, and educational researchers to support empirical investigation of music's role in social-emotional development and therapeutic applications.

Dataset Overview

The SEM Dataset provides a standardized resource for researchers investigating the relationship between musical structure and psychological impact, with potential applications in music therapy, educational technology, and affective computing.

Property Value
Total Tracks 419
Total Duration 3.5 hours
Audio Format WAV
Sample Rate 44.1 kHz
Total Size 2.5 GB
License CC BY-NC 4.0

Categorical Framework

The three-category framework reflects established constructs in social-emotional learning theory:

Category Tracks Duration Theoretical Foundation
Outlook 147 1.23 hours Promotes optimistic cognition through melodic coherence, tonal stability, and harmonic consonance
Problem-Solving 138 1.15 hours Enhances analytical thinking through structural complexity, tension-resolution patterns, and cognitive engagement
Empathetic Perspective-Taking 134 1.12 hours Facilitates perspective-taking through polyphonic textures, voice leading, and interpersonal musical dialogue

Methodological Framework

Music Selection Criteria

Track selection was guided by established theories in music cognition and educational psychology. Each category represents specific psychological mechanisms:

Outlook: Based on research linking melodic fluency and tonal stability to mood enhancement. Selected tracks demonstrate characteristics associated with stress reduction and positive affect, including:

  • Predictable melodic patterns that reduce cognitive load
  • Consonant harmonic progressions that create emotional stability
  • Moderate tempo ranges conducive to relaxation responses

Problem-Solving: Grounded in theories of musical expectation and cognitive flexibility. These tracks feature structural elements that mirror problem-solving processes:

  • Clear presentation of musical "problems" (tensions, dissonances)
  • Logical developmental processes leading to resolution
  • Balanced complexity that engages without overwhelming cognitive resources

Empathetic Perspective-Taking: Informed by research on musical perspective-taking and social cognition. Selection emphasized:

  • Polyphonic textures requiring attention to multiple musical "voices"
  • Interactive musical elements suggesting dialogue or conversation
  • Emotional narrative arcs that model empathetic responses

Research Applications

The SEM Dataset supports multiple research paradigms:

Primary Applications

  • Classification Research: Automatic categorization of music by social-emotional impact
  • Music Therapy Studies: Evidence-based music selection for therapeutic interventions
  • Educational Technology: Intelligent music recommendation for social-emotional learning
  • Cognitive Science: Investigation of music-emotion relationships

Citation and Usage

If you use the SEM Dataset in your research, please cite:

@dataset{sem_music_dataset_2025,
  title={SEM Music Dataset: A Dataset for Social-Emotional Music Classification},
  author={[Authors]},
  year={2025},
  publisher={[Publisher]},
  version={1.0},
  doi={[DOI_NUMBER]}
}

Acknowledgments

This research was supported by [Funding Sources]. We acknowledge the contributions of domain experts in music psychology and education who participated in the annotation process, as well as [Institution] for computational resources and infrastructure support.

The development of this dataset reflects collaborative efforts across disciplines, demonstrating the value of interdisciplinary approaches to complex research questions at the intersection of music, psychology, and technology.

Contact

If you have any questions regarding the SEM Music Dataset or related research, please feel free to contact:

📧 Prof. Shanken — shanken@cycu.edu.tw

About

This dataset maps multiple positive emotional elements to musical features from classical and New Age music. Validated by experts, it supports emotion recognition, social-emotional learning, and therapeutic music applications, especially for special needs populations.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •