Actor Collaboration Graphs from IMDb Action and Romance Films

1,000 ego-networks represent actors and actresses, with edges connecting those who appeared in the same movie. Graphs-datasets compiled this dataset for graph classification tasks, last updated in February 2023. The graphs are derived from movies in the Action and Romance genres on IMDb.

Use Cases

• Classify ego-networks as belonging to the Action or Romance genre based on graph structure.
• Analyze collaboration patterns and network properties like clustering coefficients within actor graphs.
• Benchmark graph embedding or kernel methods on a dataset of 1,000 social networks.

Strengths

• Contains 1,000 distinct graphs for model training and evaluation.
• Graphs are derived from a well-known, public source (IMDb).

Limitations

• Dataset size is relatively small with only 1,000 graphs, which may limit model generalization.
• Limited to only two movie genres (Action and Romance), reducing diversity of collaboration patterns.
• Specific graph metrics like node count per graph or edge density are not provided in the summary.

Provenance

Source

IMDb (Internet Movie Database).

Collection Method

Ego-networks were constructed for actors/actresses, with edges for co-appearance in a movie.

Freshness

Last updated 2023-02-07.

Geography

Global, based on IMDb content (Region:us tag suggests potential US bias).