Identification of Speakers in Novels

Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics, pages 1312–1320, Sofia, Bulgaria, August 4-9 2013. c⃝2013 Association for Computational Linguistics Identification of Speakers in Novels Hua He† Denilson Barbosa ‡ Grzegorz Kondrak‡ †Department of Computer Science ‡Department of Computing Science University of Maryland University of Alberta huah@cs.umd.edu {denilson,gkondrak}@ualberta.ca Abstract Speaker identification is the task of at- tributing utterances to characters in a lit- erary narrative. It is challenging to auto- mate because the speakers of the majority of utterances are not explicitly identified in novels. In this paper, we present a super- vised machine learning approach for the task that incorporates several novel fea- tures. The experimental results show that our method is more accurate and general than previous approaches to the problem. 1 Introduction Novels are important as social communication documents, in which novelists develop the plot by means of discourse between various charac- ters. In spite of a frequently expressed opinion that all novels are simply variations of a certain number of basic plots (Tobias, 2012), every novel has a unique plot (or several plots) and a different set of characters. The interactions among charac- ters, especially in the form of conversations, help the readers construct a mental model of the plot and the changing relationships between charac- ters. Many of the complexities of interpersonal re- lationships, such as romantic interests, family ties, and rivalries, are conveyed by utterances. A precondition for understanding the relation- ship between characters and plot development in a novel is the identification of speakers behind all utterances. However, the majority of utterances are not explicitly tagged with speaker names, as is the case in stage plays and film scripts. In most cases, authors rely instead on the readers’ compre- hension of the story and of the differences between characters. Since manual annotation of novels is costly, a system for automatically determining speakers of utterances would facilitate other tasks related to the processing of literary texts. Speaker identifica- tion could also be applied on its own, for instance in generating high quality audio books without hu- man lectors, where each character would be iden- tifiable by a distinct way of speaking. In addi- tion, research on spoken language processing for broadcast and multi-party meetings (Salamin et al., 2010; Favre et al., 2009) has demonstrated that the analysis of dialogues is useful for the study of social interactions. In this paper, we investigate the task of speaker identification in novels. Departing from previous approaches, we develop a general system that can be trained on relatively small annotated data sets, and subsequently applied to other novels for which no annotation is available. Since every novel has its own set of characters, speaker identification cannot be formulated as a straightforward tagging problem with a universal set of fixed tags. Instead, we adopt a ranking approach, which enables our model to be applied to literary texts that are differ- ent from the ones it has been trained on. Our approach is grounded in a …

Content truncated to 500 words