Datasets:

Jaiminshahh
/

music-universe-hkg

Error code:   StreamingRowsError
Exception:    CastError
Message:      Couldn't cast
source: string
target: string
relation: string
role: string
degree_distribution: struct<5: int64, 3: int64, 7: int64, 4: int64, 9: int64, 6: int64, 11: int64, 8: int64, 10: int64, 1 (... 119 chars omitted)
  child 0, 5: int64
  child 1, 3: int64
  child 2, 7: int64
  child 3, 4: int64
  child 4, 9: int64
  child 5, 6: int64
  child 6, 11: int64
  child 7, 8: int64
  child 8, 10: int64
  child 9, 13: int64
  child 10, 17: int64
  child 11, 15: int64
  child 12, 12: int64
  child 13, 14: int64
  child 14, 16: int64
  child 15, 19: int64
  child 16, 20: int64
  child 17, 21: int64
  child 18, 18: int64
  child 19, 22: int64
communities: list<item: struct<community_id: int64, size: int64, top_members: list<item: string>, dominant_genres (... 997 chars omitted)
  child 0, item: struct<community_id: int64, size: int64, top_members: list<item: string>, dominant_genres: struct<la (... 985 chars omitted)
      child 0, community_id: int64
      child 1, size: int64
      child 2, top_members: list<item: string>
          child 0, item: string
      child 3, dominant_genres: struct<latin: int64, dancehall: int64, dance: int64, hip-hop: int64, latino: int64, chicago-house: i (... 893 chars omitted)
          child 0, latin: int64
          child 1, dancehall: int64
          child 2, dance: int64
          child 3, hip-hop: int64
          child 4, latino: int64
          child 5, chicago-house: int64
          child 6, detroit-techno: int64
          child 7, minimal-techno: int64
    
...
 name: string
top_betweenness: list<item: struct<node_id: string, betweenness: double, name: string>>
  child 0, item: struct<node_id: string, betweenness: double, name: string>
      child 0, node_id: string
      child 1, betweenness: double
      child 2, name: string
entity_type_counts: struct<Track: int64, Artist: int64, Album: int64, Genre: int64, Era: int64, AudioCluster: int64>
  child 0, Track: int64
  child 1, Artist: int64
  child 2, Album: int64
  child 3, Genre: int64
  child 4, Era: int64
  child 5, AudioCluster: int64
num_weakly_connected_components: int64
relation_type_counts: struct<PERFORMED_BY: int64, APPEARS_ON: int64, HAS_GENRE: int64, IN_AUDIO_CLUSTER: int64, POPULARITY (... 210 chars omitted)
  child 0, PERFORMED_BY: int64
  child 1, APPEARS_ON: int64
  child 2, HAS_GENRE: int64
  child 3, IN_AUDIO_CLUSTER: int64
  child 4, POPULARITY_TIER: int64
  child 5, ARTIST_GENRE: int64
  child 6, SIMILAR_AUDIO: int64
  child 7, FEATURED_ON: int64
  child 8, COLLABORATED_WITH: int64
  child 9, GENRE_CROSSOVER: int64
  child 10, RELEASED_BY: int64
  child 11, ALBUM_GENRE: int64
  child 12, GENRE_RELATED: int64
  child 13, CLUSTER_GENRE: int64
top_hubs: list<item: struct<node_id: string, degree: int64, entity_type: string, name: string>>
  child 0, item: struct<node_id: string, degree: int64, entity_type: string, name: string>
      child 0, node_id: string
      child 1, degree: int64
      child 2, entity_type: string
      child 3, name: string
max_degree: int64
to
{'entity_type_counts': {'Track': Value('int64'), 'Artist': Value('int64'), 'Album': Value('int64'), 'Genre': Value('int64'), 'Era': Value('int64'), 'AudioCluster': Value('int64')}, 'relation_type_counts': {'PERFORMED_BY': Value('int64'), 'APPEARS_ON': Value('int64'), 'HAS_GENRE': Value('int64'), 'IN_AUDIO_CLUSTER': Value('int64'), 'POPULARITY_TIER': Value('int64'), 'ARTIST_GENRE': Value('int64'), 'SIMILAR_AUDIO': Value('int64'), 'FEATURED_ON': Value('int64'), 'COLLABORATED_WITH': Value('int64'), 'GENRE_CROSSOVER': Value('int64'), 'RELEASED_BY': Value('int64'), 'ALBUM_GENRE': Value('int64'), 'GENRE_RELATED': Value('int64'), 'CLUSTER_GENRE': Value('int64')}, 'total_nodes': Value('int64'), 'total_edges': Value('int64'), 'density': Value('float64'), 'avg_degree': Value('float64'), 'max_degree': Value('int64'), 'degree_distribution': {'5': Value('int64'), '3': Value('int64'), '7': Value('int64'), '4': Value('int64'), '9': Value('int64'), '6': Value('int64'), '11': Value('int64'), '8': Value('int64'), '10': Value('int64'), '13': Value('int64'), '17': Value('int64'), '15': Value('int64'), '12': Value('int64'), '14': Value('int64'), '16': Value('int64'), '19': Value('int64'), '20': Value('int64'), '21': Value('int64'), '18': Value('int64'), '22': Value('int64')}, 'top_hubs': List({'node_id': Value('string'), 'degree': Value('int64'), 'entity_type': Value('string'), 'name': Value('string')}), 'num_artist_communities': Value('int64'), 'communities': List({'community_id': Value('int64')
...
64'), 'progressive-house': Value('int64'), 'edm': Value('int64'), 'trance': Value('int64'), 'happy': Value('int64'), 'drum-and-bass': Value('int64'), 'hardstyle': Value('int64'), 'breakbeat': Value('int64'), 'pop-film': Value('int64'), 'indian': Value('int64'), 'k-pop': Value('int64'), 'folk': Value('int64'), 'party': Value('int64'), 'j-idol': Value('int64'), 'power-pop': Value('int64'), 'alt-rock': Value('int64'), 'punk-rock': Value('int64'), 'cantopop': Value('int64'), 'acoustic': Value('int64'), 'mandopop': Value('int64'), 'show-tunes': Value('int64'), 'dub': Value('int64'), 'heavy-metal': Value('int64'), 'metalcore': Value('int64'), 'death-metal': Value('int64'), 'dubstep': Value('int64'), 'pagode': Value('int64'), 'forro': Value('int64'), 'sertanejo': Value('int64'), 'samba': Value('int64'), 'mpb': Value('int64'), 'funk': Value('int64'), 'brazil': Value('int64'), 'j-dance': Value('int64'), 'turkish': Value('int64'), 'world-music': Value('int64'), 'gospel': Value('int64'), 'groove': Value('int64'), 'salsa': Value('int64'), 'honky-tonk': Value('int64'), 'children': Value('int64'), 'kids': Value('int64'), 'afrobeat': Value('int64'), 'tango': Value('int64'), 'jazz': Value('int64'), 'blues': Value('int64')}}), 'top_betweenness': List({'node_id': Value('string'), 'betweenness': Value('float64'), 'name': Value('string')}), 'top_pagerank': List({'node_id': Value('string'), 'pagerank': Value('float64'), 'name': Value('string')}), 'num_weakly_connected_components': Value('int64')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 99, in get_rows_or_raise
                  return get_rows(
                         ^^^^^^^^^
                File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
                  return func(*args, **kwargs)
                         ^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/src/worker/utils.py", line 77, in get_rows
                  rows_plus_one = list(itertools.islice(ds, rows_max_number + 1))
                                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2690, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2227, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2251, in _iter_arrow
                  for key, pa_table in self.ex_iterable._iter_arrow():
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 494, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 384, in _iter_arrow
                  for key, pa_table in self.generate_tables_fn(**gen_kwags):
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 295, in _generate_tables
                  self._cast_table(pa_table, json_field_paths=json_field_paths),
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 128, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2321, in table_cast
                  return cast_table_to_schema(table, schema)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2249, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              source: string
              target: string
              relation: string
              role: string
              degree_distribution: struct<5: int64, 3: int64, 7: int64, 4: int64, 9: int64, 6: int64, 11: int64, 8: int64, 10: int64, 1 (... 119 chars omitted)
                child 0, 5: int64
                child 1, 3: int64
                child 2, 7: int64
                child 3, 4: int64
                child 4, 9: int64
                child 5, 6: int64
                child 6, 11: int64
                child 7, 8: int64
                child 8, 10: int64
                child 9, 13: int64
                child 10, 17: int64
                child 11, 15: int64
                child 12, 12: int64
                child 13, 14: int64
                child 14, 16: int64
                child 15, 19: int64
                child 16, 20: int64
                child 17, 21: int64
                child 18, 18: int64
                child 19, 22: int64
              communities: list<item: struct<community_id: int64, size: int64, top_members: list<item: string>, dominant_genres (... 997 chars omitted)
                child 0, item: struct<community_id: int64, size: int64, top_members: list<item: string>, dominant_genres: struct<la (... 985 chars omitted)
                    child 0, community_id: int64
                    child 1, size: int64
                    child 2, top_members: list<item: string>
                        child 0, item: string
                    child 3, dominant_genres: struct<latin: int64, dancehall: int64, dance: int64, hip-hop: int64, latino: int64, chicago-house: i (... 893 chars omitted)
                        child 0, latin: int64
                        child 1, dancehall: int64
                        child 2, dance: int64
                        child 3, hip-hop: int64
                        child 4, latino: int64
                        child 5, chicago-house: int64
                        child 6, detroit-techno: int64
                        child 7, minimal-techno: int64
                  
              ...
               name: string
              top_betweenness: list<item: struct<node_id: string, betweenness: double, name: string>>
                child 0, item: struct<node_id: string, betweenness: double, name: string>
                    child 0, node_id: string
                    child 1, betweenness: double
                    child 2, name: string
              entity_type_counts: struct<Track: int64, Artist: int64, Album: int64, Genre: int64, Era: int64, AudioCluster: int64>
                child 0, Track: int64
                child 1, Artist: int64
                child 2, Album: int64
                child 3, Genre: int64
                child 4, Era: int64
                child 5, AudioCluster: int64
              num_weakly_connected_components: int64
              relation_type_counts: struct<PERFORMED_BY: int64, APPEARS_ON: int64, HAS_GENRE: int64, IN_AUDIO_CLUSTER: int64, POPULARITY (... 210 chars omitted)
                child 0, PERFORMED_BY: int64
                child 1, APPEARS_ON: int64
                child 2, HAS_GENRE: int64
                child 3, IN_AUDIO_CLUSTER: int64
                child 4, POPULARITY_TIER: int64
                child 5, ARTIST_GENRE: int64
                child 6, SIMILAR_AUDIO: int64
                child 7, FEATURED_ON: int64
                child 8, COLLABORATED_WITH: int64
                child 9, GENRE_CROSSOVER: int64
                child 10, RELEASED_BY: int64
                child 11, ALBUM_GENRE: int64
                child 12, GENRE_RELATED: int64
                child 13, CLUSTER_GENRE: int64
              top_hubs: list<item: struct<node_id: string, degree: int64, entity_type: string, name: string>>
                child 0, item: struct<node_id: string, degree: int64, entity_type: string, name: string>
                    child 0, node_id: string
                    child 1, degree: int64
                    child 2, entity_type: string
                    child 3, name: string
              max_degree: int64
              to
              {'entity_type_counts': {'Track': Value('int64'), 'Artist': Value('int64'), 'Album': Value('int64'), 'Genre': Value('int64'), 'Era': Value('int64'), 'AudioCluster': Value('int64')}, 'relation_type_counts': {'PERFORMED_BY': Value('int64'), 'APPEARS_ON': Value('int64'), 'HAS_GENRE': Value('int64'), 'IN_AUDIO_CLUSTER': Value('int64'), 'POPULARITY_TIER': Value('int64'), 'ARTIST_GENRE': Value('int64'), 'SIMILAR_AUDIO': Value('int64'), 'FEATURED_ON': Value('int64'), 'COLLABORATED_WITH': Value('int64'), 'GENRE_CROSSOVER': Value('int64'), 'RELEASED_BY': Value('int64'), 'ALBUM_GENRE': Value('int64'), 'GENRE_RELATED': Value('int64'), 'CLUSTER_GENRE': Value('int64')}, 'total_nodes': Value('int64'), 'total_edges': Value('int64'), 'density': Value('float64'), 'avg_degree': Value('float64'), 'max_degree': Value('int64'), 'degree_distribution': {'5': Value('int64'), '3': Value('int64'), '7': Value('int64'), '4': Value('int64'), '9': Value('int64'), '6': Value('int64'), '11': Value('int64'), '8': Value('int64'), '10': Value('int64'), '13': Value('int64'), '17': Value('int64'), '15': Value('int64'), '12': Value('int64'), '14': Value('int64'), '16': Value('int64'), '19': Value('int64'), '20': Value('int64'), '21': Value('int64'), '18': Value('int64'), '22': Value('int64')}, 'top_hubs': List({'node_id': Value('string'), 'degree': Value('int64'), 'entity_type': Value('string'), 'name': Value('string')}), 'num_artist_communities': Value('int64'), 'communities': List({'community_id': Value('int64')
              ...
              64'), 'progressive-house': Value('int64'), 'edm': Value('int64'), 'trance': Value('int64'), 'happy': Value('int64'), 'drum-and-bass': Value('int64'), 'hardstyle': Value('int64'), 'breakbeat': Value('int64'), 'pop-film': Value('int64'), 'indian': Value('int64'), 'k-pop': Value('int64'), 'folk': Value('int64'), 'party': Value('int64'), 'j-idol': Value('int64'), 'power-pop': Value('int64'), 'alt-rock': Value('int64'), 'punk-rock': Value('int64'), 'cantopop': Value('int64'), 'acoustic': Value('int64'), 'mandopop': Value('int64'), 'show-tunes': Value('int64'), 'dub': Value('int64'), 'heavy-metal': Value('int64'), 'metalcore': Value('int64'), 'death-metal': Value('int64'), 'dubstep': Value('int64'), 'pagode': Value('int64'), 'forro': Value('int64'), 'sertanejo': Value('int64'), 'samba': Value('int64'), 'mpb': Value('int64'), 'funk': Value('int64'), 'brazil': Value('int64'), 'j-dance': Value('int64'), 'turkish': Value('int64'), 'world-music': Value('int64'), 'gospel': Value('int64'), 'groove': Value('int64'), 'salsa': Value('int64'), 'honky-tonk': Value('int64'), 'children': Value('int64'), 'kids': Value('int64'), 'afrobeat': Value('int64'), 'tango': Value('int64'), 'jazz': Value('int64'), 'blues': Value('int64')}}), 'top_betweenness': List({'node_id': Value('string'), 'betweenness': Value('float64'), 'name': Value('string')}), 'top_pagerank': List({'node_id': Value('string'), 'pagerank': Value('float64'), 'name': Value('string')}), 'num_weakly_connected_components': Value('int64')}
              because column names don't match

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

🎵 Music Universe — Heterogeneous Knowledge Graph (HKG)

A richly structured Heterogeneous Knowledge Graph mapping the interconnected music universe with 166,330 nodes, 759,236 edges, 6 entity types, and 16 relation types.

📊 Graph Statistics

Metric	Value
Total Nodes	166,330
Total Edges	759,236
Entity Types	6 (Artist, Track, Album, Genre, AudioCluster, Era)
Relation Types	16
Artist Communities	1,341 (Louvain)
Graph Density	0.000027

🏗️ Schema

Entity Types

Type	Count	Ontology Reference
🎤 Artist	29,858	`mo:MusicArtist` / MusicBrainz:Artist
🎵 Track	89,740	`mo:Track` / MusicBrainz:Recording
💿 Album	46,589	`mo:Record` / MusicBrainz:Release
🎭 Genre	113	MusicBrainz:Genre / Wikidata:P136
📊 AudioCluster	25	Derived (K-Means on audio features)
📅 Era	5	Derived (popularity tiers)

Relation Types

Relation	Source → Target	Count
PERFORMED_BY	Track → Artist	123,424
APPEARS_ON	Track → Album	89,740
HAS_GENRE	Track → Genre	89,740
IN_AUDIO_CLUSTER	Track → AudioCluster	89,740
POPULARITY_TIER	Track → Era	89,740
RELEASED_BY	Album → Artist	85,490
ALBUM_GENRE	Album → Genre	46,589
COLLABORATED_WITH	Artist ↔ Artist	38,684
SIMILAR_AUDIO	Artist ↔ Artist	37,939
FEATURED_ON	Artist → Track	33,684
ARTIST_GENRE	Artist → Genre	32,243
GENRE_CROSSOVER	Artist → Genre	2,733
GENRE_RELATED	Genre ↔ Genre	537
CLUSTER_GENRE	AudioCluster → Genre	75

📁 Files

File	Format	Description
`data/nodes.jsonl`	JSONL	All 166K nodes with properties
`data/edges.jsonl`	JSONL	All 759K edges with relation types and properties
`data/triples.tsv`	TSV	(subject, predicate, object) triple format
`data/schema.json`	JSON	Full schema with entity/relation type definitions
`data/analytics_report.json`	JSON	Graph analytics, communities, centrality
`data/graph.pkl`	Pickle	NetworkX DiGraph for direct Python loading
`data/REPORT.md`	Markdown	Comprehensive analysis report

🚀 Usage

import pickle
import networkx as nx

# Load the full graph
from huggingface_hub import hf_hub_download
path = hf_hub_download("Jaiminshahh/music-universe-hkg", "data/graph.pkl", repo_type="dataset")
with open(path, "rb") as f:
    G = pickle.load(f)

print(f"Nodes: {G.number_of_nodes()}, Edges: {G.number_of_edges()}")

# Find all collaborators of an artist
artist_node = [n for n, d in G.nodes(data=True) if d.get("name") == "Armin van Buuren"][0]
collabs = [(G.nodes[v].get("name"), d.get("track_count")) 
           for _, v, d in G.out_edges(artist_node, data=True)
           if d.get("relation") == "COLLABORATED_WITH"]

# Get genre landscape
genre_edges = [(G.nodes[u].get("name"), G.nodes[v].get("name"), d.get("jaccard_similarity"))
               for u, v, d in G.edges(data=True) if d.get("relation") == "GENRE_RELATED"]

📚 Research Foundation

WASABI Song Corpus & KG (LREC 2019, arxiv:1912.02477) — Schema design
KGAT (KDD 2019, arxiv:1905.07854) — KG recommendation patterns
CypherBench (arxiv:2412.18702) — Property graph schema methodology
Music Ontology (musicontology.com) — Ontology alignment
MusicBrainz Entity Model — Entity type definitions

🔗 Interactive Explorer

Explore this graph interactively: Music Universe HKG Explorer Space

Data Source

Built from maharshipandya/spotify-tracks-dataset with derived relationships computed via collaboration analysis, audio feature similarity (cosine), genre co-occurrence (Jaccard), and K-Means clustering.

Downloads last month: 15

Total file size:

252 MB

Space using Jaiminshahh/music-universe-hkg 1

Papers for Jaiminshahh/music-universe-hkg

CypherBench: Towards Precise Retrieval over Full-scale Modern Knowledge Graphs in the LLM Era

Paper • 2412.18702 • Published Dec 24, 2024 • 8

Love Me, Love Me, Say (and Write!) that You Love Me: Enriching the WASABI Song Corpus with Lyrics Annotations

Paper • 1912.02477 • Published Dec 5, 2019

KGAT: Knowledge Graph Attention Network for Recommendation

Paper • 1905.07854 • Published May 20, 2019