Dark Mode
Spotify Track Characteristics
Not Specified
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This dataset provides insights into musical preferences, specifically for Spotify song recommendations. It features 200 songs along with various audio statistics collected directly from Spotify's API. The primary aim was to model musical preference by distinguishing between liked and disliked tracks. It contains 100 liked songs, predominantly French Rap, American rap, rock, and electro music, alongside 95 disliked songs spanning genres such as metal, classical, and disco. This collection facilitates the development of machine learning models to understand and predict individual music taste.
Columns
The dataset comprises 14 columns, each describing a distinct audio feature or preference indicator:
- acousticness: A measure from 0.0 to 1.0 indicating the likelihood of a track being acoustic.
- danceability: A value from 0.0 to 1.0 describing how suitable a track is for dancing, based on tempo, rhythm, and beat strength.
- duration_ms: The length of the track in milliseconds.
- energy: A measure from 0.0 to 1.0 representing the intensity and activity of a track; higher values indicate faster, louder, or noisier tracks.
- instrumentalness: Predicts the absence of vocals in a track, with values closer to 1.0 suggesting no vocal content.
- key: The musical key of the track, represented by integers where 0 is C, 1 is C♯/D♭, and so forth.
- liveness: Detects the presence of an audience; higher values suggest a live performance.
- loudness: The average overall loudness of the track in decibels (dB), typically ranging from -60 to 0 dB.
- mode: Indicates the modality of the track (major represented by 1, minor by 0).
- speechiness: Detects the presence of spoken words; values above 0.66 likely mean the track is entirely speech, while values between 0.33 and 0.66 may include rap or layered speech and music.
- tempo: The estimated speed of the track in beats per minute (BPM).
- time_signature: An estimated overall time signature of the track.
- valence: A measure from 0.0 to 1.0 describing the musical positiveness conveyed by a track; high valence indicates happiness, low valence indicates sadness.
- liked: The target variable, indicating user preference (1 for liked songs, 0 for disliked songs).
Distribution
The data is provided in a CSV format and is approximately 14.37 kB in size. It contains 195 valid records across all 14 columns, consisting of 100 liked songs and 95 disliked songs. The dataset structure is tabular, with each row representing a unique song and its corresponding audio features.
Usage
This dataset is ideal for:
- Developing and testing machine learning models for music recommendation systems.
- Classifying songs based on audio features to predict user preference.
- Conducting data analysis to understand the relationships between various musical attributes and listener likes/dislikes.
- Exploring the anatomy of musical preference using objective audio characteristics.
Coverage
The dataset's scope is primarily musical genre-based, reflecting the collector's personal preferences. It includes a variety of genres, such as French Rap, American Rap, rock, electro, metal, classical, and disco. The data was collected using Spotify's API. There is no specific geographic or time range information detailed beyond the collection methodology.
License
CC0: Public Domain
Who Can Use It
This dataset is particularly useful for:
- Data scientists and machine learning engineers working on recommendation algorithms.
- Music researchers interested in the quantitative analysis of musical attributes.
- Students learning about data collection from APIs, feature engineering, and classification tasks in machine learning.
- Developers aiming to build or enhance music-related applications.
Dataset Name Suggestions
- Spotify Music Preference Features
- Audio Features for Music Recommendation
- Liked vs Disliked Songs Dataset
- Spotify Track Characteristics
- Musical Taste Prediction Data
Attributes
Original Data Source: Spotify Track Characteristics
QUALITY