Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Affect as a proxy for literary mood

Emily Öhman, Riikka Rossi

TL;DR

The paper tackles quantifying literary mood by treating affect as a proxy for mood and distinguishing mood from tone. It deploys a lexicon-based mood-detection approach using the Finnish Emotion Intensity Lexicon (FEIL), augmented to address domain-specific language and semantic shift, and applies it to a Finnish literary corpus from the late 19th and early 20th centuries. Key findings show that mood signals extracted from opening passages align with expert qualitative analyses, and that addressing language-specific morphology and semantic drift is crucial for reliable mood estimation. This work advances digital humanities by integrating affective computing with traditional literary analysis and demonstrates the potential for robust mood detection across languages and corpora using interpretable lexicon-based methods.

Abstract

We propose to use affect as a proxy for mood in literary texts. In this study, we explore the differences in computationally detecting tone versus detecting mood. Methodologically we utilize affective word embeddings to look at the affective distribution in different text segments. We also present a simple yet efficient and effective method of enhancing emotion lexicons to take both semantic shift and the domain of the text into account producing real-world congruent results closely matching both contemporary and modern qualitative analyses.

Affect as a proxy for literary mood

TL;DR

The paper tackles quantifying literary mood by treating affect as a proxy for mood and distinguishing mood from tone. It deploys a lexicon-based mood-detection approach using the Finnish Emotion Intensity Lexicon (FEIL), augmented to address domain-specific language and semantic shift, and applies it to a Finnish literary corpus from the late 19th and early 20th centuries. Key findings show that mood signals extracted from opening passages align with expert qualitative analyses, and that addressing language-specific morphology and semantic drift is crucial for reliable mood estimation. This work advances digital humanities by integrating affective computing with traditional literary analysis and demonstrates the potential for robust mood detection across languages and corpora using interpretable lexicon-based methods.

Abstract

We propose to use affect as a proxy for mood in literary texts. In this study, we explore the differences in computationally detecting tone versus detecting mood. Methodologically we utilize affective word embeddings to look at the affective distribution in different text segments. We also present a simple yet efficient and effective method of enhancing emotion lexicons to take both semantic shift and the domain of the text into account producing real-world congruent results closely matching both contemporary and modern qualitative analyses.
Paper Structure (10 sections, 3 figures, 2 tables)

This paper contains 10 sections, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Background & Previous Work
  3. Data
  4. The Emotion Intensity Lexicon
  5. Method
  6. The Mood in Selected Texts
  7. An overview of the selected texts
  8. Results
  9. Discussion
  10. Future Work

Figures (3)

  • Figure 1: Co-occurence of emotions in FEIL.
  • Figure 2: Emotion word distribution in first three paragraphs per 1000 words
  • Figure 3: Emotion word distribution in the first 200 tokens of each chapter per 1000 words