Extracting Social Networks from Literary Fiction

Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, pages 138–147, Uppsala, Sweden, 11-16 July 2010. c⃝2010 Association for Computational Linguistics Extracting Social Networks from Literary Fiction David K. Elson Dept. of Computer Science Columbia University delson@cs.columbia.edu Nicholas Dames English Department Columbia University nd122@columbia.edu Kathleen R. McKeown Dept. of Computer Science Columbia University kathy@cs.columbia.edu Abstract We present a method for extracting so- cial networks from literature, namely, nineteenth-century British novels and se- rials. We derive the networks from di- alogue interactions, and thus our method depends on the ability to determine when two characters are in conversation. Our approach involves character name chunk- ing, quoted speech attribution and conver- sation detection given the set of quotes. We extract features from the social net- works and examine their correlation with one another, as well as with metadata such as the novel’s setting. Our results provide evidence that the majority of novels in this time period do not fit two characterizations provided by literacy scholars. Instead, our results suggest an alternative explanation for differences in social networks. 1 Introduction Literary studies about the nineteenth-century British novel are often concerned with the nature of the community that surrounds the protagonist. Some theorists have suggested a relationship be- tween the size of a community and the amount of dialogue that occurs, positing that “face to face time” diminishes as the number of characters in the novel grows. Others suggest that as the social setting becomes more urbanized, the quality of di- alogue also changes, with more interactions occur- ring in rural communities than urban communities. Such claims have typically been made, however, on the basis of a few novels that are studied in depth. In this paper, we aim to determine whether an automated study of a much larger sample of nineteenth century novels supports these claims. The research presented here is concerned with the extraction of social networks from literature. We present a method to automatically construct a network based on dialogue interactions between characters in a novel. Our approach includes com- ponents for finding instances of quoted speech, attributing each quote to a character, and iden- tifying when certain characters are in conversa- tion. We then construct a network where char- acters are vertices and edges signify an amount of bilateral conversation between those charac- ters, with edge weights corresponding to the fre- quency and length of their exchanges. In contrast to previous approaches to social network construc- tion, ours relies on a novel combination of pattern- based detection, statistical methods, and adapta- tion of standard natural language tools for the liter- ary genre. We carried out this work on a corpus of 60 nineteenth-century novels and serials, includ- ing 31 authors such as Dickens, Austen and Conan Doyle. In order to evaluate the literary claims in ques- tion, we compute various characteristics of the dialogue-based social network and stratify these results by categories such as the novel’s setting. For example, the density of the network provides evidence about …

Content truncated to 500 words