摘要： Marvel is one of the many entertainment companies that is using data visualization to improve its business model in 2021.
▲圖片標題(來源：Dreamstime.com)
Last year, we talked about the growing importance of big data in the entertainment industry. Marvel is one of the many companies using big data to optimize its business model.
As we all know, Marvel is one of the most influential comic books in the world created by Stan Lee. Only a mind like his could create an out-of-this-world creation that would last forever. What’s amazing is that Marvel characters are developed through the influence of other Marvel heroes through data visualization.
The dataset of Marvel is packed with a list of manifestations of their co-superheroes. For instance, when Spider-Man appears in a comic book with Captain America, these are all visualized through data graphics.
This is essential since the Marvel Universe consists of thousands of unique universes and all the multiverse stories happened on earth. Through data visualization, they will know the heroes who are much more important than those with fewer priorities.
Better Understanding of Marvel and Its Evolution with Big Data
Marvel was first released as a comic book in October 1939 that featured a few superheroes, known as Human Torch and Sub-Mariner. In 1941, the Marvel comic released Captain America. However, the publisher of Marvel Comics named Goldman closed its door from publishing this kind of book in the early 1950s.
The reason behind this is that readers lose interest in reading this kind of comic genre. However, in 1953 they tried to bring back superhero comics to the public and continued releasing a series of books about Captain America.
As of today, many Marvel superheroes have their movies that gather more people to take an interest in Marvel. So, to know more about them, it’s best to go through the complete list of Marvel movies in order to learn about the backstories of different Marvel superheroes.
Big data has become more important than ever in Marvel’s business model. More Marvel movies and comics are being adapted based on demographic data and data visualization.
First Graphic Presentation
To visualize the network, we need to first build a small web application using Python. Then, together with sigma.js and using the networkx package, we came up with the first graphic visualization. The node’s size represents the degree and the colors used on the node to graph detected clusters.
With the help of this first graphical presentation, we can come up with a few hypotheses. This might be less interesting to you, but this graph visualization is indeed essential. Here are some of the findings:
Marvel characters who got the highest score on the social graph are Captain America, Spider-Man, and Iron Man. It means these are the most relevant heroes in the Marvel Universe. It is not a surprise anymore since they have been around the Comic series since the beginning.
Some characters who also received a high score from the social graph are Thor’s universe, The Avengers, Fantastic Four, and X-Men. Some small clusters are detected in the graph, and most of these characters are less popular, and some don’t even have a page on Wikipedia.
Thus, this graphical visualization is essential to know how they will visualize the next Marvel film that Marvel fans will be a big hit.
Shaping the Graph
Using edge weights, we can shape the internal structure of the graph. Edge weight is the number of co-occurrences between heroes. Let’s say, for example, the edge-weight connection between Spider-Man and Captain America is the same as the number of series in comics they have been together.
Thus, the more we increase the focus on a specific hero like Captain America, the graph will be much more clearer. In that way, the graph will only present the heroes with related data and connection with Captain America; heroes with less connection will disappear.
Identifying the Marvel Influencers
Through these graphs, we will clearly know how Marvel groups its superheroes in certain movies. There is broad research and has considered many statistics through social networks such as Twitter and Facebook to come up with these visualizations.
A few graph criteria were presented to explain them: Page Rank, closeness centrality, and degree centrality. It also showed that the value of closeness centrality represents the node’s importance in the graph on how close the characters are to each other and those that aren’t connected.
若喜歡本文，請關注我們的臉書 Please Like our Facebook Page： Big Data In Finance
留下你的回應
以訪客張貼回應