A Label-Aware BERT Attention Network for Zero-Shot Multi-Intent Detection in Spoken Language Understanding

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 492–497 November 7–11, 2021. c⃝2021 Association for Computational Linguistics 492 Inducing Stereotypical Character Roles from Plot Structure Labiba Jahan Department of Mathematics & Computer Science Augustana College 639 38th Street, Rock Island, IL 61201 USA labibajahan@augustana.edu Rahul Mittal, Mark A. Finlayson Knight Foundation School of Computing and Information Sciences Florida International University 11200 S.W. 8th Street, Miami, FL 33199 USA {rmitt008, markaf}@fiu.edu Abstract Stereotypical character roles—also known as archetypes or dramatis personae—play an im- portant function in narratives: they facilitate efficient communication with bundles of de- fault characteristics and associations and ease understanding of those characters’ roles in the overall narrative. We present a fully unsuper- vised k-means clustering approach for learn- ing stereotypical roles given only structural plot information. We demonstrate the tech- nique on Vladimir Propp’s structural theory of Russian folktales (captured in the extended ProppLearner corpus, with 46 tales), showing that our approach can induce six out of seven of Propp’s dramatis personae with F1 mea- sures of up to 0.70 (0.58 average), with an additional category for minor characters. We have explored various feature sets and varia- tions of a cluster evaluation method. The best- performing feature set comprises plot func- tions, unigrams, tf-idf weights, and embed- dings over coreference chain heads. Roles that are mentioned more often (Hero, Villain), or have clearly distinct plot patterns (Princess) are more strongly differentiated than less fre- quent or distinct roles (Dispatcher, Helper, Donor). Detailed error analysis suggests that the quality of the coreference chain and plot functions annotations are critical for this task. We provide all our data and code for repro- ducibility1. 1 Introduction Stereotypical characters are characters that both play an important role in the plot of a story and fit into recognizable categories. In general, characters are central to every narrative and drive the action forward, and stereotypical character roles include both common, context-independent roles such as Hero, Villain, or Victim, as well as culturally- specific roles such as the Donor (in, for example, 1https://doi.org/10.34703/gzx1-9v95/ DD6SEN Russian tales) or the Trickster (in, for example, Na- tive American tales). Referred to alternatively as archetypes (Abrams and Harpham, 2014) or drama- tis personae (Propp, 1968), stereotypical character roles are crucial aids to narrative understanding: they facilitate efficient communication with bun- dles of default characteristics and associations and ease understanding of the purpose of those char- acter in the overall narrative (Robbins, 2005). Be- yond demonstrated cognitive effects, stereotypical character roles are useful for NLP tasks such as narrative generation (Gervás, 2013), interactive dia- logue generation (Rowe et al., 2008), and sentiment analysis (Bhaskaran and Bhallamudi, 2019). Prior work has demonstrate the utility of pre- identified roles. But how do we learn the roles in the first place? There have been several approaches to this task, but all prior work incorporated some a priori knowledge of the possible stereotypical roles in the model, for example, results of manual qualitative analyses (Harun and Jamaludin, 2016), an …

Content truncated to 500 words