Abstract
Marvel Entertainment movie production house has rolled out movies that have seen record commercial success over the years. In this article, the authors investigate the social network formed by connecting the characters that have featured in the Marvel movies and thus examine their structural topology. Characters like Quicksilver and Hulk emerge as the main “bridging” characters. Movies, such as the Avengers series, emerge as central “bridging” movies. A positive relationship between user ratings and commercial success was seen with larger star cast being preferred choice of the audience in the recent years – especially the ones that bring supermen together. The network of 393 actors consists of 2814 unique edges. Cluster analysis using the Clauset-Newman-Moore cluster algorithm detects communities of actors who have acted together, but yet may be connected to the overall character network through a character who may have co-acted in two different movies dominated by the different “universes.”
TopBackground
Extant literature has looked at several aspects of implementing social network analysis on character interactions. Ding and Yilmaz (2010) used statistical learning to estimate the affinity of characters in movies. Social Network analysis then identified leaders in the communities they formed. Weng, Chu, and Wu (2009) investigated the perspective of social relationships in movies. Using a specially designed method called RoleNet, the authors extract relationships and construct role’s social network, thereby leading to identification of corresponding communities. The study was further able to prove the superiority of ‘social-based story segmentation approach’ over other conventional methods. Using global face name matching Zhang, Xu, Lu, and Huang (2009) identified characters in films and the relationship between characters were mined by applying Social network analysis. A recent study by Lv, Wu, Zhu, and Wang (2018) used a model called as StoryRoleNet to determine relationship among roles. Their study also analysed networks constructed using video and subtitle text.
Key Terms in this Chapter
Social Network: A social network is a connection between any two or more entities joined together through a relationship.
Cinematic Universe: Universe that has significant presence in films and television.
Degree: Degree represents the number of direct connections of a node.
Network Density: Density of a network is the ratio between the actual connections in the network in ratio to the maximum number of possible connections. Determines how well connected or sparsely connected a network is.
Network Communities: Communities in a network are the clump of densely connected nodes in a network. Communities are detected through cluster algorithms.
Structural Holes: Structural holes are the absence of connections in the node’s immediate neighbourhood.
Betweenness Centrality: Betweenness centrality is path-based that determines how much a node forms a ‘bridge’ between other nodes in a network. Those with high betweenness centrality typically control the flow of resources in a network.