Characters are fundamental to literary analysis. Current approaches are heavily reliant on NER to identify characters, causing many to be overlooked. We propose a novel technique for character detection, achieving significant improvements over state of the art on multiple datasets.

Anaphor resolution is an important task in NLP with many applications. Despite much research effort, it remains an open problem. The difficulty of the problem varies substantially across different sub-problems. One sub-problem, in particular, has been largely untouched by prior work despite occurring frequently throughout corpora: the anaphor that has multiple antecedents, which here we call multi-antecedent anaphors or manaphors. Current coreference resolvers restrict anaphors to at most a single antecedent. As we show in this paper, relaxing this constraint poses serious problems in coreference chain-building, where each chain is intended to refer to a single entity. This work provides a formalization of the new task with preliminary insights into multi-antecedent noun-phrase anaphors, and offers a method for resolving such cases that outperforms a number of baseline methods by a significant margin. Our system uses local agglomerative clustering on candidate antecedents and an existing coreference system to score clusters to determine which cluster of mentions is antecedent for a given anaphor. When we augment an existing coreference system with our proposed method, we observe a substantial increase in performance (0.6 absolute CoNLL F1) on an annotated corpus.

Across a variety of cultural fields, researchers have identified a near ubiquitous underrepresentation and decentralization of women. This occurs both at the level of who is able to produce cultural works and who is depicted within them. Women are less likely to be directors of Hollywood films and also less likely to have starring roles. 1 Women writers are less likely to be reviewed in major book review outlets as well as do the reviewing. 2 Women are considerably less likely to

What would it mean for a novel to turn us as we turn its pages? How are we not simply moved, but transformed—turned around —through the novel’s combination of gestural and affective structures? How might we think, in other words, about the correspondences between the novel’s technics and its tropes in its ability to assume meaning for us as a genre on a profound personal level? This essay explores the use of computational models to understand the novel’s relationship to the narration of profound change as a vehicle of readerly devotion. It aims to give us new techniques for thinking about the novel’s significance as a genre, a significance that depends less on forms of critical estrangement and more on felt experiences of pronounced transformation. In doing so, it outlines the potential impact that the practice of computational modeling might have on our own affective attachments as critical readers.

No abstract