Literary style research has always relied on researchers’ subjective perceptions and qualitative descriptions, lacking objective quantitative evidence. In order to make up for this methodological limitation, this study introduced natural language processing technology to conduct a cross-field quantitative comparative analysis of the literary styles of Dickens and Austen, two representative writers. The study constructed a digital corpus of representative works of the two writers, and systematically extracted quantitative features from four dimensions: vocabulary richness, syntactic complexity, emotional distribution and topic model. Through methods such as word frequency statistics, keyword analysis, syntactic analysis, sentiment calculation, and LDA topic modeling, the systematic differences in language use between the two writers are revealed in a data-driven manner. This study aims to use accurate data to verify the core assertions about the styles of the two writers in traditional literary criticism, thereby proving the effectiveness and scientificity of the NLP measurement method as a useful supplement to traditional literary research, and providing an empirical path that can be used as a reference for literary criticism from the perspective of digital humanities.

Bingxian Chen, Jingjing Wu. Literary Style Quantitative Analysis Based on NLP: Taking Dickens and Austen as Examples. Academic Journal of Computing & Information Science (2026), Vol. 9, Issue 6: 55-73. https://doi.org/10.25236/AJCIS.2026.090608.

[1] Zhu Benjun, Nie Hua. Cross-border and integration: Digital Humanities from a Global Perspective—A Review of the First Peking University “Digital Humanities Forum” Conference [J]. Journal of University Libraries, 2016, 34(5): 16-21.

[2] Li Yiyun, Niu Yunping. Exploring the reasons for differences in the English translation of "The Diamond Sutra" from the perspective of metrological stylistics [J]. Foreign Language Teaching, 2024, 45(2): 91-98.

[3] Zeng Yiping, Zhu Xiaowen. Application of computational methods in the study of Chinese stylistics[J]. Journal of Fujian Normal University (Philosophy and Social Sciences Edition), 2006(1): 14-17.

[4] Fan Wenhao, Zhao Yuxiang, Li Jinhao, et al. Method evolution of computational grounded theory and its application in digital humanities research [J]. Library Forum, 2026, 46(1): 11-23.

[5] Liu Ying, Xiao Tianjiu. Research on metrological style of "Dream of Red Mansions" [J]. Journal of Dream of Red Mansions, 2014(4): 260-281.

[6] Liu Ying, Xiao Tianjiu. Research on the quantitative style of Jin Yong and Gu Long’s novels [J]. Journal of Tsinghua University (Philosophy and Social Sciences Edition), 2014, 29(5): 135-147.

[7] Xiao Tianjiu, Liu Ying. Analysis of Jin Yong and Gu Long’s novel style based on clustering and classification [J]. Journal of Chinese Information Science, 2015, 29(5): 167-177.

[8] Zhang Jiayi, Li Liangbo. Research on the style of "A Moveable Feast" from the perspective of metrological stylistics [J]. Xinchu Culture, 2025(27): 27-29.

[9] Wu Jingjing, Xu Juanjuan. Research on the evaluation of the English translation of "The Song of Burial Flowers" [J]. Foreign Language Studies, 2021, 7(3): 78-87.

[10] Fu Shouxiang. "A Tale of Two Cities": Conscience and the Spirit of Benevolence of the Times [N]. Chinese Science Journal, 2019-07-03(003).

[11] Hui Haifeng. The emotional style of Austen’s novels: Research based on the corpus framework of British novels from 1768 to 1830 [J]. Journal of Guangdong University of Foreign Studies, 2026(1): 149-160.