Natural Language Processing-Based Literary Analysis Framework for Linguistic Nuance Detection in English Literature

The present paper introduces a Natural Language Processing (NLP)-Based Literary Analysis Framework that works to identify the linguistic elements in English literature. The framework leverages state-of-the-art Contextual Word Embeddings in BERT (Bidirectional Encoder Representations of Transformers), a state-of-the-art NLP model to learn the semantic nuances in literary works. Using BERT that allows identifying contextual meanings of individual words relying on the text around it, the framework can extract such complex linguistic features as irony, metaphors, and emotions in narratives. This method enables us to identify sophistication in language used important phenomena that cannot be identified in the conventional literary analysis, thus a more insightful interpretation of the language and style of an author. Implementation of BERT means that we can identify literary devices, emotional changes and thematic developments within a text which are incredibly important in actually interpreting the text beyond the generally accepted interpretations. Through the marriage of computers and linguistics this framework finds a piece of the puzzle as far as literary analysis. This framework presents a new way of viewing English literature through modern NLP techniques.