Quantitative Analysis of Style in Mihai Eminescu's Poetry

  • A Briciu Department of Computer Science, Faculty of Mathematics and Computer Science, Babes-Bolyai University, Cluj-Napoca, Romania


Quantitative stylistic methods aim to express certain aspects of a text in numeric form, thus allowing the introduction of fast, powerful and accurate computational approaches for analysis. While in the case of literature, the validity and usefulness of such studies is highly controversial, one cannot deny the opportunities brought forward by computational methods: first, the exploration of large sets of documents in search of patterns otherwise difficult to discover by human readers; second, the possibility of opening up new perspectives by uncovering latent features of texts. In this study, we investigate the poetic work of one of the most important Romanian poets, Mihai Eminescu, through a variety of quantitative methods addressing lexical, morphological, semantic and emotional aspects of text. We propose a comparison between the results of the computational approach and established interpretations of Eminescu’s work in order to assess the viability of computational methods in poetic style studies.


[1] T. Boroș, S. D. Dumitrescu, and R. Burtica. NLP-cube: End-to-end raw text processing with neural networks. In Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, pages 171–179, Brussels, Belgium, October 2018. Association for Computational Linguistics.
[2] A. Briciu and M. Lupea. RoEmolex - a Romanian Emotion Lexicon. Studia Universitatis Babeș-Bolyai Informatica, 62(2):45–56, 2017.
[3] J. F. Burrows. Computation into criticism: A study of Jane Austen’s novels and an experiment in method. Clarendon Press, 1987.
[4] M. Dascalu, D. Gˆıfu, and S. Trausan-Matu. What makes your writing style unique? Significant differences between two famous Romanian orators. In International Conference on Computational Collective Intelligence, pages 143–152. Springer International Publishing, 2016.
[5] R. Delmonte. Computing poetry style. In Proceeding of Emotion and Sentiment in Social and Expressive Media: Approaches and perspectives from AI (ESSEM 2013), CEUR Workshop, pages 148–155, 2013.
[6] S. D. Dumitrescu, A. M. Avram, L. Morogan, and S.-A. Toma. RoWordNet – a Python API for the Romanian WordNet. In 2018 10th International Conference on Electronics, Computers and Artificial Intelligence (ECAI), pages 1–6. IEEE, 2018.
[7] L. G´aldi. Stilul poetic al lui Mihai Eminescu. Editura Academiei Republicii Populare Romˆane, 1964.
[8] A. Hammond, J. Brooke, and G. Hirst. A tale of two cultures: Bringing literary analysis and computational linguistics together. In Proceedings of the Workshop on Computational Linguistics for Literature, pages 1–8, 2013.
[9] D. Irimia. Limbajul poetic eminescian. Junimea, 1979.
[10] J. T. Kao and D. Jurafsky. A computational analysis of poetic style. In LiLT (Linguistic Issues in Language Technology), volume 12, pages 1–33, 2015.
[11] D. M. Kaplan and D. M. Blei. A computational approach to style in American poetry. In Seventh IEEE International Conference on Data Mining (ICDM 2007), pages 553–558. IEEE, 2007.
[12] V. Kesarwani. Automatic Poetry Classification Using Natural Language Processing. PhD thesis, University of Ottawa, 2018.
[13] M. Kestemont and L. Herman. Can machines read (literature)? Umanistica Digitale, 3(5), 2019.
[14] M. G. Kirschenbaum. The remaking of reading: Data mining and the digital humanities. In The National Science Foundation Symposium on Next Generation of Data Mining and Cyber-Enabled Discovery for Innovation, Baltimore, MD, 2007.
[15] M. Lupea and A. Briciu. Formal Concept Analysis of a Romanian Emotion Lexicon. In 13th IEEE International Conference on Intelligent Computer Communication and Processing (ICCP), pages 111–118, 2017.
[16] M. Lupea and A. Briciu. Studying emotions in Romanian words using Formal Concept Analysis. Computer Speech & Language, 57:128 – 145, 2019.
[17] L. Meneses and R. Furuta. Visualizing poetry: Tools for critical analysis. The Journal of the Initiative for Digital Humanities, Media, and Culture, 3(1):1–14.
[18] I. E. Petrescu. Studii eminesciene. Casa Cărții de Știință, 2007.
[19] R. Simionescu. Romanian deep noun phrase chunking using Graphical Grammar Studio. In Proceedings of the 8th International Conference Linguistic Resources And Tools For Processing Of The Romanian Language, pages 135–143, 2012.
[20] D. Tatar, M. Lupea, E. Kapetanios, and G. Altmann. Hreb-like analysis of Eminescu’s poems. Glottometrics, 28:37–56, 2014.
[21] D. Tufis, and V. Barbu Mititelu. The Lexical Ontology for Romanian, volume 48 of Text, Speech and Language Technology, pages 491–504. Springer, 2014.
[22] L. Zhang and J. Gao. A comparative study to understanding about poetics based on natural language processing. Open Journal of Modern Linguistics, 7:229–237, 2017.
How to Cite
BRICIU, A. Quantitative Analysis of Style in Mihai Eminescu's Poetry. Studia Universitatis Babeș-Bolyai Informatica, [S.l.], v. 64, n. 2, p. 80-95, dec. 2019. ISSN 2065-9601. Available at: <http://www.cs.ubbcluj.ro/~studia-i/journal/journal/article/view/45>. Date accessed: 29 nov. 2020. doi: https://doi.org/10.24193/subbi.2019.2.06.