Digital Humanities and the Evolution of Digital Text Analysis in Literary Studies

This article provides a comprehensive evaluation of the historical development, methodological frameworks, and contemporary paradigms of digital text analysis within Digital Humanities (DH). Tracing the disciplinary trajectory from early mid-twentieth-century humanities computing – focused on concordance building, lexicography, and foundational stylometry – to the current era of computational literary studies driven by Natural Language Processing (NLP), deep learning, and Large Language Models (LLMs), this paper demonstrates how computational methods reshape literary scholarship. Key analytical frameworks, including distant reading, macro-analysis, sentiment analysis, social network analysis, and co-reference resolution, are systematically reviewed. Furthermore, the study addresses critical epistemological and technical challenges, such as the risk of hermeneutic reductionism, algorithmic bias, and the cross-cultural gap in applying computational tools to low-resource languages. Ultimately, the paper argues for a symbiotic hybrid methodology that bridges computational scale with qualitative close reading, offering a holistic paradigm for future literary research.