Page MenuHomePhabricator

movie and TV connections finder
Closed, ResolvedPublicFeature

Description

Recently I watched a TV show where a lot of actors have collaborated together in other TV shows or movies before. It'd be cool to have a tool to exploit this data easily from Wikidata. A few questions this tool could answer:

  • For actor X, which other actor Y has been in more than one TV shows or movies with them?
  • Which TV shows or movies have actor X and Y been in together before?
  • For TV show or movie X and Y, which actors have been in both of them?
  • What are the biggest clusters of actors who star together a lot?

This could be expanded to showrunners, producers, etc.

Event Timeline

It should not be too difficult to have an RStudio Shiny app developed for this.

Tech stack:

  • a bit of Pyspark/R for the clustering part, and then probably
  • plain SPARQL against WDQS for everything else directly from the dashboard.
Aklapper changed the subtype of this task from "Task" to "Feature Request".

A Proposed solution is built during the WMA-Hackathon-2025 to be linked to the task description

For large data sets it is better to use QLever, WDQS falls short

Bovimacoco changed the status of subtask T390402: Secure: Migrate hardcoded URLs to .env with validation from Open to In Progress.