Exploring a movies dataset with BeGraph
Who doesn’t love Cinema? For most people Cinema is present of a daily basis on their lives and plays an important role in their leisure and happiness. In this post we are going to visualize the movies network used by Livio Bioglio and Ruggero G. Pensa using BeGraph. They studied and ranked movies, actors and directors using elements of Network Science. They used a network where nodes are movies and a directed link exists between two movies when one of them has referenced/cited the other in some way. Directors do it all the time: they pay homage to their favorite movies using similar scenes, include Easter Eggs or simply talking about it in an interview. This relation defines a very interesting network on a subset of around 47000 movies in the Internet Movie DataBase. Let us see two examples in detail:
- There is a classic Easter Egg in Riders of the Lost Ark (1981) that references Star Wars: A New Hope (1977). Check the picture below and you can see a hieroglyph representing R2D2 and C3PO on the left column while Dr. Jones is looking for the Ark. Then there will be a directed link form Raiders of the Lost Ark and A New Hope in the network.
- The Odessa Steps is a mythical scene from Battleship Potemkin (1928) that has inspired many other scenes in other movies, like The Godfather (1972) or The Untouchables (1987).
Movies with high InDegree (number of incoming links) are thus the influencers of this network because they are an inspiration to many movies.
Thanks to the work by Biogio and Pensa, the movie reference network is available and the data can be exported in csv format. This is perfect for BeGraph! So, we have loaded, expanded an analyzed the network and prepared some visualizations in the BeGraph Gallery:
This network is really funny and curious to explore. In the default view of the network, node size represents the impact or influence of that movie to others. Those movies are usually Hollywood classics, like Psycho, The Wizard of Oz, Star Wars, Jaws, The Godfather, Citizen Kane …
The network presents a very big central cluster with most of the movies. We can see (as expected) that most of the movies are American, but many of the smaller clusters are formed by movies from another countries. Try to find a cluster of U.K. movies directed by Gerald Thomas and you will see how densely connected it is.
We have prepared several visualizations emphasising the release year, the genre or the country of the movie. Additionally we have calculated the Page Rank (Google’s centrality measure for webpages) for every movie, and this indicates that The Wizard of Oz is the most influencing movie of history.
We hope you enjoy this network as much as we do!!!