A Bayesian Mixed Effects Model of Literary Character

By David Bamman, Ted Underwood, Noah A. Smith. 2014

229 citations recorded by the source corpus.

We consider the problem of automatically inferring latent character types in a collection of 15,099 English novels published between 1700 and 1899. Unlike prior work in which character types are assumed responsible for probabilistically generating all text associated with a character, we introduce a model that employs multiple effects to account for the influence of extra-linguistic information (such as author). In an empirical evaluation, we find that this method leads to improved agreement with the preregistered judgments of a literary scholar, complementing the results of alternative models.

Subfields: Artificial Intelligence, Natural Language Processing, Probabilistic & Bayesian Methods, Social Sciences & Humanities

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Mohit Bansal
2. Yejin Choi
3. Graham Neubig
4. Luke S. Zettlemoyer
5. Mirella Lapata
6. Jianfeng Gao
7. Richard Socher
8. Hannaneh Hajishirzi

In Natural Language Processing

1. Mohit Bansal
2. Mirella Lapata
3. Yejin Choi
4. Phil Blunsom
5. Graham Neubig
6. Richard Socher
7. Luke S. Zettlemoyer
8. Yoav Goldberg

In Probabilistic & Bayesian Methods

1. Haoruo Peng
2. Snigdha Chaturvedi
3. Mirella Lapata
4. Li Lucy
5. Philip John Gorinski
6. Lea Frermann
7. György Szarvas
8. Hal Daumé Iii

In Social Sciences & Humanities

1. Snigdha Chaturvedi
2. Li Lucy
3. Mirella Lapata
4. Philip John Gorinski
5. Lea Frermann
6. György Szarvas
7. Hal Daumé Iii
8. Mohit Iyyer

Citations in the corpus

• The NarrativeQA Reading Comprehension Challenge (2017, background): “These include high level plot understanding through clustering of novels (Frermann and Szarvas, 2017) or summarization of movie scripts (Gorinski and Lapata, 2015), to more fine grained processing by inducing character types (Bamman et al., 2014b; Bamman et al., 2014a), understanding relationships b”
• An annotated dataset of literary entities (2019, background): “Much of this work relies on the ability to extract entities accurately, including work focused on modeling (Bamman et al., 2014; Iyyer et al., 2016; Chaturvedi et al., 2017).”
• Movie Script Summarization as Graph-based Scene Extraction (2015, methodology): “Other work (Bamman et al., 2013, 2014) automatically infers latent character types (e.g., villains or heroes) in novels and movie plot summaries.”
• Story Comprehension for Predicting What Happens Next (2017, methodology): “ing such as identifying character personas (Bamman et al., 2014; Valls-Vargas et al., 2015), interpersonal relationships (Chaturvedi, 2016), plotpatterns (Jockers, 2013), narrative structures (Finlayson, 2012).”
• Gender and Representation Bias in GPT-3 Generated Stories (2021, methodology): “We use BookNLP to find main characters and sentences containing them (Bamman et al., 2014).”
• Unsupervised Learning of Evolving Relationships Between Literary Characters (2017, methodology): “We preprocessed the narratives using the BookNLP pipeline (Bamman, Underwood, and Smith 2014) to obtain POS tags, dependency parses, and co-referent mentions, and to identify major characters.”
• Inducing Semantic Micro-Clusters from Deep Multi-View Representations of Novels (2017, methodology): “Unsupervised models have been proposed which, given raw text, extract prototypical event structure (McIntyre and Lapata, 2010; Chambers and Jurafsky, 2009), prototypical characters (Bamman et al., 2013, 2014; Elsner, 2012) and their social networks (Elson et al., 2010).”
• Generating Character Descriptions for Automatic Summarization of Fiction (2019, methodology): “Character mentions were extracted and dependency parsing was performed using BookNLP (Bamman, Underwood, and Smith 2014).”