Clustering of Novels Represented as Social Networks

Mariona Coll Adanay, Caroline Sporleder


Abstract
Within the field of literary analysis, there are few branches as confusing as that of genre theory. Literary criticism has failed so far to reach a consensus on what makes a genre a genre. In this paper, we examine the degree to which the character structure of a novel is indicative of the genre it belongs to. With the premise that novels are societies in miniature, we build static and dynamic social networks of characters as a strategy to represent the narrative structure of novels in a quantifiable manner. For each of the novels, we compute a vector of literary-motivated features extracted from their network representation. We perform clustering on the vectors and analyze the resulting clusters in terms of genre and authorship.
Anthology ID:
2015.lilt-12.4
Volume:
Linguistic Issues in Language Technology, Volume 12, 2015 - Literature Lifts up Computational Linguistics
Month:
Oct
Year:
2015
Address:
Venue:
LILT
SIG:
Publisher:
CSLI Publications
Note:
Pages:
Language:
URL:
https://www.aclweb.org/anthology/2015.lilt-12.4
DOI:
Bib Export formats:
BibTeX MODS XML EndNote
PDF:
http://aclanthology.lst.uni-saarland.de/2015.lilt-12.4.pdf