Personality Proﬁling of Fictional Characters using Sense-Level Links between Lexical Resources - paper

Sep 15, 2026

Uploaded by Stanley

The study investigates automated personality prediction for protagonists in novels using the Five-Factor Model. It introduces and publishes a collaboratively built dataset of fictional character personality and frames the task as text classification. Semantic features are derived from WordNet and VerbNet sense-level information together with word vector representations. Three machine-learning models are evaluated from speech, actions, and predicatives, with lexical-semantic features outperforming baselines most strongly, matching known psychology findings.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Introduction

Personality proﬁling as NLP task

Related work

Content-based recommendation and semantic information

Entity-centric and latent character models

Personality Proﬁling of Fictional Characters using Sense-Level Links between Lexical Resources Lucie Flekovay and Iryna Gurevychyz y Ubiquitous Knowledge Processing Lab (UKP-TUDA) Department of Computer Science, Technische Universitt Darmstadt z Ubiquitous Knowledge Processing Lab (UKP-DIPF) German Institute for Educational Research [www.ukp.tu-darmstadt.de](www.ukp.tu-darmstadt.de) “Always be yourself, unless you can be Batman. Then always be Batman.” – Bill Murray Abstract This study focuses on personality prediction of protagonists in novels based on the Five-Factor Model of personality. We present and publish a novel collaboratively built dataset of ﬁctional character personality and design our task as a text classiﬁcation problem. We incorporate a range of semantic features, including WordNet and VerbNet sense-level information and word vector representations. We evaluate three machine learning models based on the speech, actions and predicatives of the main characters, and show that especially the lexical-semantic features signiﬁcantly outperform the baselines. The most predictive features correspond to reported ﬁndings in personality psychology. 1 Introduction Recent progress in NLP has given rise to the ﬁeld of personality proﬁling-automated classiﬁcation of personality traits based on written, verbal and multimodal behavior of an individual. This research builds upon ﬁndings from classical personality psychology and has applications in a wide range of areas from medicine (suicide prevention) across security (forensics, paedophile detection, cyberbullying) to marketing and sales (recommendation systems, target group proﬁles) . The gold standard labels for an objective evaluation of personality are mostly obtained by means of personality tests of the Five Factor Model (FFM) (McCrae and Costa, 1987; Goldberg, 1990), which is wellknown and widely accepted in psychology and other research ﬁelds. The FFM deﬁnes personality along ﬁve bipolar scales: Extraversion (sociable vs. reserved), Emotional stability (secure vs. neurotic), Agreeableness (friendly vs. unsympathic), Conscientiousness (organized vs. careless) and Openness to experience (insightful vs. unimaginative) . Psychologists have shown that these ﬁve personality traits are stable across individual lifespan, demographical and cultural differences (John and Srivastava, 1999) and affect many life aspects.(Terracciano et al., 2008; Rentfrow et al., 2011) . It has been shown that the personality traits of readers impact their literature preferences (Tirre and Dixit, 1995; Mar et al., 2009) . Psychology researchers also found that perceived similarity is predictive of interpersonal attraction (Montoya et al., 2008; Byrne, 1961; Chartrand and Bargh, 1999) . More explicitly, recent research (Kaufman and Libby, 2012) shows that readers of a narrative develop more favorable attitudes and less stereotype application towards a character, if his difference (e.g. racial) is revealed only later in the story. We therefore hypothesize that readers might have a preference for reading novels depicting ﬁctional characters that are similar to themselves. Finding a direct link between reader's and protagonist's personality traits would advance the development of content-based recommendation systems. As a ﬁrst step to explore this hypothesis further, it needs tobe determined if we are able to construct a personality proﬁle of a ﬁctional character in a similar way as it is done for humans, and which aspects of personality proﬁling can be exploited to automatize such procedure. In this paper, we open this research topic by presenting a novel collaboratively built dataset of ﬁctional character personality in Section 3, which we make available on our website.1 Framing the personality prediction as a text classiﬁcation task, we incorporate features of both lexical- 1 [https://www.ukp.tu-darmstadt.de/data/](https://www.ukp.tu-darmstadt.de/data/)[ ](https://www.ukp.t

FAQs

What personality framework does the study use for fictional character profiling?

The study uses the Five-Factor Model (FFM), treating personality as five traits along bipolar scales.

How is the personality prediction problem formulated in the paper?

It is framed as a text classification task using features extracted from the protagonists’ textual evidence.

Which semantic features provide the strongest results?

Lexical-semantic features based on sense-level information from WordNet and VerbNet, together with word vector representations, significantly outperform baselines, especially for direct speech and actions.

Sep 15, 2026

Uploaded by Stanley