The parallel universes derived from the Marvel universe become more and more perfect. The characters appearing in them are unmeasurable. There are already more than 5,000 superheroes in the universe. I have always been very interested in Marvel heroes. However，the network is too large to know all the characters relationship. After watching the movie, I like to search for the heroes network organized by other people to help me clear out various relationship. These network pictures are very clear and easy understanding for people to know the hero’s relationship.
After learning the gephi，I tried to create a network map by myself which could answer the questions
- Which hero has more relationship with others?
- How many main organizations are there in the Marvel universe? Who are them？
I collected my original Marvel hero network data from Github which is already a Gephi form file. After opening it in the software, I found this data is large and mess, so I start to find some sample and method to clean and filter them.
In this sample, the author adjusts the degree level to filter the information to make this ego graphic. In Gephi, The creators first use the Force Atlas 2 layout algorithm to cluster related nodes. And they used different colors to respond to different hero in the Marvel universe.
This is the original graphic I had when I only use modularity Class to filter the database.
Now it is a mess at all, so I used the behavior alternatives to solve the overlap problems. After select those options, it appears like this.
In this graphics, we can clearly see that in the Marvel university, there are five main organizations. The green part is the largest one which is lead by Iron man, Captain America, and Vision.
Back to the question which I had in the beginning: Which hero has more relationship with others? I changed the algorithm from ForceAtlas2 to Circular layout. This layout can show the relationship between each other more clearly. To simplify this database and create the main hero circle, I changed the degree range from 1-2189 to 800-2189. Make sure only the main characters will be shown in this graphics.
In this graphic, we could see that Invisible women and the human torch have a strong connection, as well as Spiderman and Watson. The characters Iron man, Wolverine, and Scarlet Witch have more connection with other heroes in Marvel Word.
Gephi is ideal for analyzing and visualizing complex data networks. In this project, I think the biggest difficulty is to let normal viewers understand your complex network. So some of them aim to show the overall information like I didn’t put too many details like text or tags. Simply use color to response to different sections. In the following projects, I feel like that I should improve the diversity of my data. The graphics made with only one database is not enough. If there are more related data to make different graphics and create a storyboard is more helpful for viewers to understand it.