Exploring AI-Based Sentiment, Emotion, and Tone Measurement Tools in the Interpretation of English Literature: A Comparative Framework for AI-Assisted Literary Analysis

Abstract

The increasing integration of artificial intelligence within literary scholarship has significantly expanded the methodological possibilities of digital humanities research. In particular, sentiment analysis and emotion detection models have been widely used in the analysis of large textual corpora and in the examination of affective patterns in literary narratives. Notwithstanding these leaps, deep desegregation persists in the interpretive reliability of computational strategies for the literary language of perplexity. Figurative expression, narrative voice, irony, and contextual nuances often confound automated sentiment classification systems, thereby posing important methodological questions about the role of artificial intelligence in literary interpretation. This work fills a key gap in digital humanities scholarship by presenting a Comparative AI-Assisted Literary Interpretation Framework (CAALIF) that implements a method for systematically testing interpretive results from various sentiment analysis models. The framework involves lexicon-based, transformer-based, and large language model approaches, comparing their analytical performance within a structured interpretive pipeline. The analysis is applied to a corpus of canonical English literary texts which cover a range of narrative traditions - tragedy, realist fiction and modernist literature. The results show high levels of divergence among computational models in interpreting literary sentiment, especially in passages characterised by figurative language and narrative ambiguity. While AI analytics tools can effectively identify general emotional patterns and sentiment shifts across the stages of a narrative, their interpretations offer a reflection on why the process of humanistic critical assessment remains critical. The creation by this study of a consciousness of a comparative methodological framework for artificial intelligence (AI)-assisted literary analysis contributes to a continuous debate regarding the integration of computational methodologies with literary studies. It provides a liberal strategy for enhancing methodological rigour in digital humanities research.