💁‍♀️Please take a moment to carefully read through this description and metadata to better understand the dataset and its nuances before proceeding to the Suggestions and Discussions section.

Dataset Description:

This dataset compiles the tracks from Spotify's official "Top Tracks of 2023" playlist, showcasing the most popular and influential music of the year according to Spotify's streaming data. It represents a wide range array of genres, artists, and musical styles that have defined the musical landscapes of 2023. Each track in the dataset is detailed with a variety of features, popularity, and metadata. This dataset serves as an excellent resource for music enthusiasts, data analysts, and researchers aiming to explore music trends or develop music recommendation systems based on empirical data.

Data Collection and Processing:

Obtaining the Data:

The data was obtained directly from the Spotify Web API, specifically from the "Top Tracks of 2023" official playlist curated by Spotify. The Spotify API provides detailed information about tracks, artists, and albums through various endpoints.

Data Processing:

To process and structure the data, I developed Python scripts using data science libraries such as pandas for data manipulation and spotipy for API interactions specifically for Spotify data retrieval.

Workflow:

1. Authentification
2. API Requests
3. Data Cleaning and Transformation
4. Saving the Data

Attribute Descriptions:

• artist_name: the artist name
• track_name: the title of the track
• is_explicit: Indicates whether the track contains explicit content
• album_release_date: The date when the track was released
• genres: A list of genres associated with the track's artist(s)
• danceability: A measure from 0.0 to 1.0 indicating how suitable a track is for dancing based on a combination of musical elements
• valence: A measure from 0.0 to 1.0 indicating the musical positiveness conveyed by a track
• energy: A measure from 0.0 to 1.0 representing a perceptual measure of intensity and activity
• loudness: The overall loudness of a track in decibels (dB)
• acousticness: A measure from 0.0 to 1.0 whether the track is acoustic.
• instrumentalness: Predicts whether a track contains no vocals
• liveness: Detects the presence of an audience in the recordings
• speechiness: Detects the presence of spoken words in a track
• key: The key the track is in. Integers map to pitches using standard Pitch Class notation.
• tempo: The overall estimated tempo of a track in beats per minute (BPM)
• mode: Modality of the track
• duration_ms: The length of the track in milliseconds
• time_signature: An estimated overall time signature of a track
• popularity: A score between 0 and 100, with 100 being the most popular

Possible Data Projects

• Trends Analysis
• Genre Popularity
• Mood and Music
• Comparison with other tracks

Disclaimer and Responsible Use:

• This dataset, derived from Spotify's "Top Tracks of 2023" playlist, is intended for educational, research, and analysis purposes only. Users are urged to use this data responsibly and ethically.
• Users should comply with Spotify's Terms of Service and Developer Policies when using this dataset.
• The dataset includes music track information such as names and artist details, which are subject to copyright. While the dataset presents this information for analytical purposes, it does not convey any rights to the music itself.
• Users of the dataset must ensure that their use does not infringe on the rights of copyright holders. Any analysis, distribution, or derivative work should respect the intellectual property rights of all parties and comply with applicable laws.
• The dataset is provided "as is," without warranty, and the creator disclaims any legal liability for the use of the dataset by others. Users are responsible for ensuring their use of the dataset is legal and ethical.
• For the most accurate and up-to-date information regarding Spotify's music, playlists, and policies, users are encouraged to refer directly to Spotify's official website. This ensures that users have access to the latest details directly from the source.
• The creator/maintainer of this dataset is not affiliated with Spotify, any third-party entities, or artists mentioned within the dataset. This project is independent and has not been authorized, sponsored, or otherwise approved by Spotify or any other mentioned entities.

Contribution

I encourage users who discover new insights, propose dataset enhancements, or craft analytics that illuminate aspects of the dataset's focus to share their findings with the community. - Kaggle Notebooks: To facilitate sharing and collaboration, users are encouraged to create and share their analyses through Kaggle notebooks. For ease of use, start your notebook by clicking "New Notebook" atop this dataset’s page on K...

Content

This is a dataset of Spotify tracks over a range of 125 different genres. Each track has some audio features associated with it. The data is in CSV format which is tabular and can be loaded quickly.

  Usage

The dataset can be used for:

Building a Recommendation System based on some user input or preference Classification purposes based on audio features and available genres Any other application that you can think of. Feel free to discuss!

  Column… See the full description on the dataset page: https://huggingface.co/datasets/maharshipandya/spotify-tracks-dataset.

How many paid subscribers does Spotify have? As of the first quarter of 2025, Spotify had 268 million premium subscribers worldwide, up from 239 million in the corresponding quarter of 2024. Spotify’s subscriber base has increased dramatically in the last few years and has more than doubled since early 2019. Spotify and competitors Spotify is a music streaming service originally founded in 2006 in Sweden. The platform can be used from various devices and allows users to browse through a catalogue of music licensed through multiple record labels, as well as creating and sharing playlists with other users. Additionally, listeners are able to enjoy music for free with advertisements or are also given the option to purchase a subscription to allow for unlimited ad-free music streaming. Spotify’s largest competitors are Pandora, a company that offers a similar service and remains popular in the United States, and Apple Music, which was launched in 2015. While Pandora was once among the highest-grossing music apps in the Apple App Store, recent rankings show that global services like QQ Music, NetEase Cloud Music, and YouTube Music now generate higher monthly revenues.Users are also able to register Spotify accounts using Facebook directly through the website using an app. This enables them to connect with other Facebook friends and explore their music tastes and playlists. Spotify is a popular source for keeping up-to-date with music, and the ability to enjoy Spotify anywhere at any time allows consumers to shape their music consumption around their lifestyles and preferences.

This dataset is based on the subset of users in the #nowplaying dataset who publish their #nowplaying tweets via Spotify. In principle, the dataset holds users, their playlists and the tracks contained in these playlists.

The csv-file holding the dataset contains the following columns: "user_id", "artistname", "trackname", "playlistname", where

user_id is a hash of the user's Spotify user name

artistname is the name of the artist

trackname is the title of the track and

playlistname is the name of the playlist that contains this track.

The separator used is , each entry is enclosed by double quotes and the escape character used is .

A description of the generation of the dataset and the dataset itself can be found in the following paper:

Pichl, Martin; Zangerle, Eva; Specht, Günther: "Towards a Context-Aware Music Recommendation Approach: What is Hidden in the Playlist Name?" in 15th IEEE International Conference on Data Mining Workshops (ICDM 2015), pp. 1360-1365, IEEE, Atlantic City, 2015.

This dataset was created using Spotify developer API. It consists of user-created as well as Spotify-curated playlists.
The dataset consists of 1 million playlists, 3 million unique tracks, 3 million unique albums, and 1.3 million artists.
The data is stored in a SQL database, with the primary entities being songs, albums, artists, and playlists.
Each of the aforementioned entities are represented by unique IDs (Spotify URI).
Data is stored into following tables:

• album
• artist
• track
• playlist
• track_artist1
• track_playlist1

album

| id | name | uri |

id: Album ID as provided by Spotify
name: Album Name as provided by Spotify
uri: Album URI as provided by Spotify

artist

| id | name | uri |

id: Artist ID as provided by Spotify
name: Artist Name as provided by Spotify
uri: Artist URI as provided by Spotify

track

| id | name | duration | popularity | explicit | preview_url | uri | album_id |

id: Track ID as provided by Spotify
name: Track Name as provided by Spotify
duration: Track Duration (in milliseconds) as provided by Spotify
popularity: Track Popularity as provided by Spotify
explicit: Whether the track has explicit lyrics or not. (true or false)
preview_url: A link to a 30 second preview (MP3 format) of the track. Can be null
uri: Track Uri as provided by Spotify
album_id: Album Id to which the track belongs

playlist

| id | name | followers | uri | total_tracks |

id: Playlist ID as provided by Spotify
name: Playlist Name as provided by Spotify
followers: Playlist Followers as provided by Spotify
uri: Playlist Uri as provided by Spotify
total_tracks: Total number of tracks in the playlist.

track_artist1

| track_id | artist_id |

Track-Artist association table

track_playlist1

| track_id | playlist_id |

Track-Playlist association table

- - - - - SETUP - - - - -

The data is in the form of a SQL dump. The download size is about 10 GB, and the database populated from it comes out to about 35GB.

spotifydbdumpschemashare.sql contains the schema for the database (for reference):
spotifydbdumpshare.sql is the actual data dump.

Setup steps:
1. Create database

- - - - - PAPER - - - - -

The description of this dataset can be found in the following paper:

Papreja P., Venkateswara H., Panchanathan S. (2020) Representation, Exploration and Recommendation of Playlists. In: Cellier P., Driessens K. (eds) Machine Learning and Knowledge Discovery in Databases. ECML PKDD 2019. Communications in Computer and Information Science, vol 1168. Springer, Cham

This dataset is a processed version of a popular Spotify tracks dataset. It includes musical features such as energy, valence, danceability, acousticness, and instrumentalness for each track, along with a newly added user preference score. The user_preference column represents a continuous value indicating how much a user likes each track, making this dataset ideal for regression tasks. The original dataset contained over 100,000 tracks; this version contains 9,999 labeled entries with user preferences added. The goal of this dataset is to enable analysis of how musical features influence user tastes, support building predictive models, and demonstrate practical applications of linear regression and other regression-based machine learning techniques. This dataset is well-suited for educational purposes, music analytics, and building simple recommendation systems.

Spotify Million Playlist Dataset Challenge Summary The Spotify Million Playlist Dataset Challenge consists of a dataset and evaluation to enable research in music recommendations. It is a continuation of the RecSys Challenge 2018, which ran from January to July 2018. The dataset contains 1,000,000 playlists, including playlist titles and track titles, created by users on the Spotify platform between January 2010 and October 2017. The evaluation task is automatic playlist continuation: given a seed playlist title and/or initial set of tracks in a playlist, to predict the subsequent tracks in that playlist. This is an open-ended challenge intended to encourage research in music recommendations, and no prizes will be awarded (other than bragging rights). Background Playlists like Today’s Top Hits and RapCaviar have millions of loyal followers, while Discover Weekly and Daily Mix are just a couple of our personalized playlists made especially to match your unique musical tastes. Our users love playlists too. In fact, the Digital Music Alliance, in their 2018 Annual Music Report, state that 54% of consumers say that playlists are replacing albums in their listening habits. But our users don’t love just listening to playlists, they also love creating them. To date, over 4 billion playlists have been created and shared by Spotify users. People create playlists for all sorts of reasons: some playlists group together music categorically (e.g., by genre, artist, year, or city), by mood, theme, or occasion (e.g., romantic, sad, holiday), or for a particular purpose (e.g., focus, workout). Some playlists are even made to land a dream job, or to send a message to someone special. The other thing we love here at Spotify is playlist research. By learning from the playlists that people create, we can learn all sorts of things about the deep relationship between people and music. Why do certain songs go together? What is the difference between “Beach Vibes” and “Forest Vibes”? And what words do people use to describe which playlists? By learning more about nature of playlists, we may also be able to suggest other tracks that a listener would enjoy in the context of a given playlist. This can make playlist creation easier, and ultimately help people find more of the music they love. Dataset To enable this type of research at scale, in 2018 we sponsored the RecSys Challenge 2018, which introduced the Million Playlist Dataset (MPD) to the research community. Sampled from the over 4 billion public playlists on Spotify, this dataset of 1 million playlists consist of over 2 million unique tracks by nearly 300,000 artists, and represents the largest public dataset of music playlists in the world. The dataset includes public playlists created by US Spotify users between January 2010 and November 2017. The challenge ran from January to July 2018, and received 1,467 submissions from 410 teams. A summary of the challenge and the top scoring submissions was published in the ACM Transactions on Intelligent Systems and Technology. In September 2020, we re-released the dataset as an open-ended challenge on AIcrowd.com. The dataset can now be downloaded by registered participants from the Resources page. Each playlist in the MPD contains a playlist title, the track list (including track IDs and metadata), and other metadata fields (last edit time, number of playlist edits, and more). All data is anonymized to protect user privacy. Playlists are sampled with some randomization, are manually filtered for playlist quality and to remove offensive content, and have some dithering and fictitious tracks added to them. As such, the dataset is not representative of the true distribution of playlists on the Spotify platform, and must not be interpreted as such in any research or analysis performed on the dataset. Dataset Contains 1000 examples of each scenario: Title only (no tracks) Title and first track Title and first 5 tracks First 5 tracks only Title and first 10 tracks First 10 tracks only Title and first 25 tracks Title and 25 random tracks Title and first 100 tracks Title and 100 random tracks Download Link Full Details: https://www.aicrowd.com/challenges/spotify-million-playlist-dataset-challenge Download Link: https://www.aicrowd.com/challenges/spotify-million-playlist-dataset-challenge/dataset_files {"references": ["C.W. Chen, P. Lamere, M. Schedl, and H. Zamani. Recsys Challenge 2018: Automatic Music Playlist Continuation. In Proceedings of the 12th ACM Conference on Recommender Systems (RecSys '18), 2018."]}

This dataset provides insights into user music listening behavior on the Spotify platform. It captures key information about individual music streaming sessions, including:

Track Information:

• spotify_track_uri: A unique identifier for each track on Spotify.
• track_name: The name of the track.
• artist_name: The name of the artist who performed the track.
• album_name: The name of the album the track belongs to.

Playback Details:

• ts: Timestamp indicating when the playback of the track ended.
• platform: The device or platform used to stream the track (e.g., mobile, desktop, web player).
• ms_played: The duration of the track played in milliseconds.

Playback Behavior:

• reason_start: The reason why the track started playing (e.g., user selection, autoplay, recommendation).
• reason_end: The reason why the track stopped playing (e.g., natural end, user skip, interruption).
• shuffle: Indicates whether shuffle mode was active during playback (TRUE or FALSE).
• skipped: Indicates whether the user manually skipped to the next track (TRUE or FALSE).

Potential Uses

This dataset can be valuable for various analyses

• Understanding user listening habits: Identifying popular tracks, artists, and albums; analyzing listening patterns across different platforms and times of day; understanding the impact of shuffle mode on listening behavior.
• Improving music recommendations: Analyzing user skips and playback durations to refine recommendation algorithms.
• Evaluating platform performance: Identifying and addressing issues related to playback quality and interruptions.
• Developing personalized music experiences: Tailoring features and recommendations based on individual listening preferences.

Content

This is a dataset of Spotify tracks over a range of 125 different genres. Each track has some audio features associated with it. The data is in CSV format which is tabular and can be loaded quickly.

Usage

The dataset can be used for:

• Building a Recommendation System based on some user input or preference
• Classification purposes based on audio features and available genres
• Any other application that you can think of. Feel free to discuss!

Column Description

• track_id: The Spotify ID for the track
• artists: The artists' names who performed the track. If there is more than one artist, they are separated by a ;
• album_name: The album name in which the track appears
• track_name: Name of the track
• popularity: The popularity of a track is a value between 0 and 100, with 100 being the most popular. The popularity is calculated by algorithm and is based, in the most part, on the total number of plays the track has had and how recent those plays are. Generally speaking, songs that are being played a lot now will have a higher popularity than songs that were played a lot in the past. Duplicate tracks (e.g. the same track from a single and an album) are rated independently. Artist and album popularity is derived mathematically from track popularity.
• duration_ms: The track length in milliseconds
• explicit: Whether or not the track has explicit lyrics (true = yes it does; false = no it does not OR unknown)
• danceability: Danceability describes how suitable a track is for dancing based on a combination of musical elements including tempo, rhythm stability, beat strength, and overall regularity. A value of 0.0 is least danceable and 1.0 is most danceable
• energy: Energy is a measure from 0.0 to 1.0 and represents a perceptual measure of intensity and activity. Typically, energetic tracks feel fast, loud, and noisy. For example, death metal has high energy, while a Bach prelude scores low on the scale
• key: The key the track is in. Integers map to pitches using standard Pitch Class notation. E.g. 0 = C, 1 = C♯/D♭, 2 = D, and so on. If no key was detected, the value is -1
• loudness: The overall loudness of a track in decibels (dB)
• mode: Mode indicates the modality (major or minor) of a track, the type of scale from which its melodic content is derived. Major is represented by 1 and minor is 0
• speechiness: Speechiness detects the presence of spoken words in a track. The more exclusively speech-like the recording (e.g. talk show, audio book, poetry), the closer to 1.0 the attribute value. Values above 0.66 describe tracks that are probably made entirely of spoken words. Values between 0.33 and 0.66 describe tracks that may contain both music and speech, either in sections or layered, including such cases as rap music. Values below 0.33 most likely represent music and other non-speech-like tracks
• acousticness: A confidence measure from 0.0 to 1.0 of whether the track is acoustic. 1.0 represents high confidence the track is acoustic
• instrumentalness: Predicts whether a track contains no vocals. "Ooh" and "aah" sounds are treated as instrumental in this context. Rap or spoken word tracks are clearly "vocal". The closer the instrumentalness value is to 1.0, the greater likelihood the track contains no vocal content
• liveness: Detects the presence of an audience in the recording. Higher liveness values represent an increased probability that the track was performed live. A value above 0.8 provides strong likelihood that the track is live
• valence: A measure from 0.0 to 1.0 describing the musical positiveness conveyed by a track. Tracks with high valence sound more positive (e.g. happy, cheerful, euphoric), while tracks with low valence sound more negative (e.g. sad, depressed, angry)
• tempo: The overall estimated tempo of a track in beats per minute (BPM). In musical terminology, tempo is the speed or pace of a given piece and derives directly from the average beat duration
• time_signature: An estimated time signature. The time signature (meter) is a notational convention to specify how many beats are in each bar (or measure). The time signature ranges from 3 to 7 indicating time signatures of 3/4, to 7/4.
• track_genre: The genre in which the track belongs

Acknowledgement

Image credits: BPR world

These are the published date of music videos of every song in

https://www.kaggle.com/edumucelli/spotifys-worldwide-daily-song-ranking

Most of the time, music videos published dates are same as music themselves.

It would be valid to use the dates as release dates.

There are no other sources better than youtube to cover as much songs as possible.

• The file contains no header

• 20 songs remained Nan (unavailable to find related videos)

• This data was retrieved by Youtube API

Context

This dataset consists of 24000 tracks from 30 genres, and is a shrunk version of maharshipandya/spotify-tracks-dataset dataset. All non-heuristic data is cut and cleaned for better usability and performance. All data taken from Spotify API and is open source. This dataset can be used to train prediction models based on user preferences, or categorise tracks by corresponding heuristic.

  Column Description

danceability: Danceability describes how suitable a track is… See the full description on the dataset page: https://huggingface.co/datasets/engels/spotify-tracks-lite.

👍 If this dataset was useful to you, leave your vote at the top of the page 👍

The dataset provides information on the daily top 200 tracks listened to by users of the Spotify digital platform around the world.

I put together this dataset because I really love music (I listen to it for several hours a day) and have not found a similar dataset with track genres on kaggle.

The dataset can be useful for beginners in the field of working with data. It contains missing values, arrays in columns, and so on, which can be great practice when conducting an EDA phase.

Soon, my example will appear here as possible, based on the specified dataset, go on a musical journey around the world and understand how the musical tastes of humanity have changed around the world)))

In addition, I will be very happy to see the work of the community on this dataset.

Also, in case of interest in data by country, I am ready to place it upon request.

You can contact me through: telegram @natarov_ivan

MGD: Music Genre Dataset

Over recent years, the world has seen a dramatic change in the way people consume music, moving from physical records to streaming services. Since 2017, such services have become the main source of revenue within the global recorded music market. Therefore, this dataset is built by using data from Spotify. It provides a weekly chart of the 200 most streamed songs for each country and territory it is present, as well as an aggregated global chart.

Considering that countries behave differently when it comes to musical tastes, we use chart data from global and regional markets from January 2017 to December 2019, considering eight of the top 10 music markets according to IFPI: United States (1st), Japan (2nd), United Kingdom (3rd), Germany (4th), France (5th), Canada (8th), Australia (9th), and Brazil (10th).

We also provide information about the hit songs and artists present in the charts, such as all collaborating artists within a song (since the charts only provide the main ones) and their respective genres, which is the core of this work. MGD also provides data about musical collaboration, as we build collaboration networks based on artist partnerships in hit songs. Therefore, this dataset contains:

Genre Networks: Success-based genre collaboration networks

Genre Mapping: Genre mapping from Spotify genres to super-genres

Artist Networks: Success-based artist collaboration networks

Artists: Some artist data

Hit Songs: Hit Song data and features

Charts: Enhanced data from Spotify Weekly Top 200 Charts

This dataset was originally built for a conference paper at ISMIR 2020. If you make use of the dataset, please also cite the following paper:

Gabriel P. Oliveira, Mariana O. Silva, Danilo B. Seufitelli, Anisio Lacerda, and Mirella M. Moro. Detecting Collaboration Profiles in Success-based Music Genre Networks. In Proceedings of the 21st International Society for Music Information Retrieval Conference (ISMIR 2020), 2020.

@inproceedings{ismir/OliveiraSSLM20, title = {Detecting Collaboration Profiles in Success-based Music Genre Networks}, author = {Gabriel P. Oliveira and Mariana O. Silva and Danilo B. Seufitelli and Anisio Lacerda and Mirella M. Moro}, booktitle = {21st International Society for Music Information Retrieval Conference} pages = {726--732}, year = {2020} }

Unlock powerful insights with our custom music datasets, offering access to millions of records from popular music platforms like Spotify, SoundCloud, Amazon Music, YouTube Music, and more. These datasets provide comprehensive data points such as track titles, artists, albums, genres, release dates, play counts, playlist details, popularity scores, user-generated tags, and much more, allowing you to analyze music trends, listener behavior, and industry patterns with precision. Use these datasets to optimize your music strategies by identifying trending tracks, analyzing artist performance, understanding playlist dynamics, and tracking audience preferences across platforms. Gain valuable insights into streaming habits, regional popularity, and emerging genres to make data-driven decisions that enhance your marketing campaigns, content creation, and audience engagement. Whether you’re a music producer, marketer, data analyst, or researcher, our music datasets empower you with the data needed to stay ahead in the ever-evolving music industry. Available in various formats such as JSON, CSV, and Parquet, and delivered via flexible options like API, S3, or email, these datasets ensure seamless integration into your workflows.

Many people used musical media via music streaming service providers to cope with the limitations of the COVID-19 pandemic. Accounting for such behavior from the perspective of uses-and-gratifications theory and situated cognition yields reliable explanations regarding people’s active and goal-oriented use of musical media. We accessed Spotify’s daily top 200 charts and their audio features from the DACH countries for the period during the first lockdown in 2020 and a comparable non-pandemic period situation in 2019 to support those theoretical explanations quantitatively with open data. After exploratory data analyses, applying a k-means clustering algorithm across the DACH countries allowed us to reduce the dimensionality of selected audio features. Following these clustering results, we discuss how these clusters are explainable using the arousal-valence-circumplex model and possibly be understood as (gratification) potentials that listeners can interact with to modulate their moods and thus emotionally cope with the stress of the pandemic. Then, we modeled a cross-validated binary SVM classifier to classify the two periods based on the extracted clusters and the remaining manifest variables (e.g., chart position) as input variables. The final test scenario of the classification task yielded high overall accuracy in classifying the periods as distinguishable classes. We conclude that these demonstrated approaches are generally suitable to classify the two periods based on the extracted mood clusters and the other input variables, and furthermore to interpret, by considering the model-related caveats, everyday music listening via those proxy variables as an emotion-focused coping strategy during the COVID-19 pandemic in DACH countries. Dataset for: Kalustian, K., & Ruth, N. (2021). “Evacuate the Dancefloor”: Exploring and Classifying Spotify Music Listening Before and During the COVID-19 Pandemic in DACH Countries. In: T. Fischinger, & C. Louven, C. (Eds.), Musikpsychologie – Empirische Forschungen - Ästhetische Experimente, Band 30.

The most successful music streaming service in the United States was Apple Music as of September, with the most up to date information showing that 49.5 million users accessed the platform each month. Spotify closely followed, with a similarly impressive 47.7 million monthly users.

What is a music streaming service?

Music streaming services provide their users with a database compiled of songs, playlists, albums and videos, where content can be accessed online, downloaded, shared, bookmarked and organized.

The music streaming business is huge, and has sometimes been lauded as the savior of the music industry. The biggest two services are in constant competition for the monopoly of the market. Apple Music was launched in 2015, whereas Spotify has been around since 2008. Other popular streaming services include Deezer, SoundCloud and iHeartRadio.

Do artists make a lot of money from streaming services? 

In short, unfortunately not. Both Apple Music and Spotify have been frequently criticized for the tiny royalty payments they offer artists. Particularly for emerging talent, streaming services are far from a lucrative source of income. Bigger, established stars like Taylor Swift are more likely to regularly make a good amount of money this way. But either way, a track needs to go viral or be streamed several million times before it earns any real cash.

A breakdown of how Spotify’s AI playlists are generated using user prompts, NLP, and music metadata.

Spotify Recommandation

( You can check how I used this dataset on my github repository )

I am basically a HUGE fan of music ( mostly French rap though with some exceptions but I love music ). And someday , while browsing stuff on Internet , I found the Spotify's API . I knew I had to use it when I found out you could get information like danceability about your favorite songs just with their id's.

https://user-images.githubusercontent.com/86613710/127216769-745ac143-7456-4464-bbe3-adc53872c133.png" alt="image">

Once I saw that , my machine learning instincts forced me to work on this project.

1. Data Collection

1.1 Playlist creation

I collected 100 liked songs and 95 disliked songs

For those I like , I made a playlist of my favorite 100 songs. It is mainly French Rap , sometimes American rap , rock or electro music.

For those I dislike , I collected songs from various kind of music so the model will have a broader view of what I don't like

There is : - 25 metal songs ( Cannibal Corps ) - 20 " I don't like " rap songs ( PNL ) - 25 classical songs - 25 Disco songs

I didn't include any Pop song because I'm kinda neutral about it

1.2 Getting the ID's

1. From the Spotify's API "Get a playlist's Items" , I turned the playlists into json formatted data which cointains the ID and the name of each track ( ids/yes.py and ids/no.py ). NB : on the website , specify "items(track(id,name))" in the fields format , to avoid being overwhelmed by useless data.

2. With a script ( ids/ids_to_data.py ) , I turned the json data into a long string with each ID separated with a comma.

1.3 Getting the statistics

Now I just had to enter the strings into the Spotify API "Get Audio Features from several tracks" and get my data files ( data/good.json and data/dislike.json )

2. Data features

From Spotify's API documentation :

• acousticness : A confidence measure from 0.0 to 1.0 of whether the track is acoustic. 1.0 represents high confidence the track is acoustic.
• danceability : Danceability describes how suitable a track is for dancing based on a combination of musical elements including tempo, rhythm stability, beat strength, and overall regularity. A value of 0.0 is least danceable and 1.0 is most danceable.
• duration_ms : The duration of the track in milliseconds.
• energy : Energy is a measure from 0.0 to 1.0 and represents a perceptual measure of intensity and activity. Typically, energetic tracks feel fast, loud, and noisy. For example, death metal has high energy, while a Bach prelude scores low on the scale. Perceptual features contributing to this attribute include dynamic range, perceived loudness, timbre, onset rate, and general entropy.
• instrumentalness : Predicts whether a track contains no vocals. “Ooh” and “aah” sounds are treated as instrumental in this context. Rap or spoken word tracks are clearly “vocal”. The closer the instrumentalness value is to 1.0, the greater likelihood the track contains no vocal content. Values above 0.5 are intended to represent instrumental tracks, but confidence is higher as the value approaches 1.0.
• key : The key the track is in. Integers map to pitches using standard Pitch Class notation . E.g. 0 = C, 1 = C♯/D♭, 2 = D, and so on.
• liveness : Detects the presence of an audience in the recording. Higher liveness values represent an increased probability that the track was performed live. A value above 0.8 provides strong likelihood that the track is live.
• loudness : The overall loudness of a track in decibels (dB). Loudness values are averaged across the entire track and are useful for comparing relative loudness of tracks. Loudness is the quality of a sound that is the primary psychological correlate of physical strength (amplitude). Values typical range between -60 and 0 db.
• mode : Mode indicates the modality (major or minor) of a track, the type of scale from which its melodic content is derived. Major is represented by 1 and minor is 0.
• speechiness : Speechiness detects the presence of spoken words in a track. The more exclusively speech-like the recording (e.g. talk s...

Estimates suggest that Apple Music had 95 million subscribers worldwide in June 2024, up by 2 million from the previous year. Launched in 2015 by U.S. tech giant Apple, Apple Music is the second largest music streaming service worldwide, competing with market leader Spotify. Spotify remains market leader While Apple Music is a popular music streaming platform, accounting for 12.6 percent of subscribers worldwide, the 2008 founded streaming service Spotify remains the market leader with a subscriber share of nearly 32 percent. Financially this meant that the Swedish company generated a global revenue of 3.7 billion euros through its Premium accounts in the fourth quarter of 2024 alone.Music streaming overall increasesOverall, music streaming has experienced significant growth over the last decade. Even if the annual growth rate is gradually declining, it still stood at over 7 percent in 2024, becoming the music industry’s main revenue driver and reaching a revenue of 20 billion U.S. dollars worldwide in 2024.

The MuSe (Music Sentiment) dataset contains sentiment information for 90,408 songs. We computed scores for the affective dimensions of valence, dominance and arousal, based on the user-generated tags that are available for each song via Last.fm. In addition, we provide artist and title metadata as well as a Spotify ID and a MusicBrainz ID, which allow researchers to extend the dataset with further metadata, such as genre or year.

Though the tags themselves cannot be included in the dataset, we include a jupyter notebook in our accompanying Github repository that demonstrates how to fetch the tags of a given song from the Last.fm API (Last.fm_API.ipynb)

We further include a jupyter notebook in the same repository that demonstrates how one might enrich the dataset with audio features using different endpoints of the Spotify API using the included Spotify IDs (spotify_API.ipynb). Please note that in its current form, the dataset only contains tentative spotify IDs for a subset (around 68%) of the songs.