Waumans Michaël C, Nicodème Thibaut, Bersini Hugues
École polytechnique de Bruxelles CoDE-IRIDIA, ULB, Brussels, Belgium.
PLoS One. 2015 Jun 3;10(6):e0126470. doi: 10.1371/journal.pone.0126470. eCollection 2015.
In a world where complex networks are an increasingly important part of science, it is interesting to question how the new reading of social realities they provide applies to our cultural background and in particular, popular culture. Are authors of successful novels able to reproduce social networks faithful to the ones found in reality? Is there any common trend connecting an author's oeuvre, or a genre of fiction? Such an analysis could provide new insight on how we, as a culture, perceive human interactions and consume media. The purpose of the work presented in this paper is to define the signature of a novel's story based on the topological analysis of its social network of characters. For this purpose, an automated tool was built that analyses the dialogs in novels, identifies characters and computes their relationships in a time-dependent manner in order to assess the network's evolution over the course of the story.
在一个复杂网络在科学中日益重要的世界里，探讨它们所提供的对社会现实的新解读如何适用于我们的文化背景，尤其是流行文化，是很有趣的。成功小说的作者能否忠实再现现实中存在的社会网络？连接一位作家的全部作品或一种小说体裁是否有共同趋势？这样的分析可以为我们作为一种文化如何看待人际互动和消费媒体提供新的见解。本文所呈现工作的目的是基于小说人物社会网络的拓扑分析来定义小说故事的特征。为此，构建了一个自动化工具，该工具分析小说中的对话，识别角色并以时间相关的方式计算他们的关系，以便评估故事过程中网络的演变。
