ccm/2025-24679-tabular-dataset

Dataset Card for ccm/2025-24679-tabular-dataset

<!-- Provide a quick summary of the dataset. --> This dataset captures self-reported music listening behaviors and preferences, including listening hours, library size, playlist creation, sharing habits, preferred decades, and social context (alone vs. with others). It was created as a class exercise in data collection and augmentation.

Dataset Details

Dataset Description

This dataset captures self-reported music listening behaviors and preferences, including listening hours, library size, playlist creation, sharing habits, preferred decades, and social context (alone vs. with others). It was created as a class exercise in data collection and augmentation.

• —Curated by: Fall 2025 24-679 course at Carnegie Mellon University
• —Shared by [optional]: Christopher McComb
• —License: MIT

Uses

<!-- Address questions around how the dataset is intended to be used. -->

Direct Use

<!-- This section describes suitable use cases for the dataset. -->

The direct intended use for this dataset is to perform classification and regressio tasks related to the labels (i.e., classifying solo/group listening preferences, regressing for hours of music listened per week)

Out-of-Scope Use

This dataset will not work well for most applications due to its size and synthetic nature. It is intended only for training.

Dataset Structure

<!-- This section provides a description of the dataset fields, and additional information about the dataset structure such as criteria used to create the splits, relationships between data points, etc. -->

This dataset contains both an original split (which contains original photographs) and an augmented split (which augmented images based on the original split).

Dataset Creation

Curation Rationale

<!-- Motivation for the creation of this dataset. --> The dataset was created to provide students with a structured but low-stakes dataset for experimenting with supervised learning workflows. Music listening was chosen because it is familiar, engaging, and relatively easy to survey.

Source Data

The source data was collected on campus at CMU

Data Collection and Processing

<!-- This section describes the data collection and processing process such as data selection criteria, filtering and normalization methods, tools and libraries used, etc. -->

Responses were collected via short in-class surveys.

Who are the source data producers?

<!-- This section describes the people or systems who originally created the data. It should also include self-reported demographic or identity information for the source data creators if this information is available. -->

• —Original data: Students in the course, who self-reported their music listening habits.
• —Augmented data: Automatically generated by course instructors and teaching assistants.

Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->

• —Small sample size: Only 36 original respondents; dataset is not representative.
• —Synthetic data: Augmented examples may not reflect real-world distributions.
• —Demographic skew: Original responses come from a limited population (engineering students at a U.S. university).

Recommendations

<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->

• —Use only for experimentation and demonstration, not for scientific or commercial conclusions.
• —Treat augmented data cautiously, especially for evaluating model generalization.
• —When teaching, emphasize dataset limitations as part of the learning process.

Dataset Card Contact

Christopher McComb (Carnegie Mellon University) — ccm@cmu.edu