Identifying Latin authors through maximum-likelihood Dirichlet inference: A contribution to model-based stylometry
DOI:
https://doi.org/10.22394/2412-9410-2021-7-1-183-198Keywords:
stylometry, LATIN literature, Dirichlet distribution, Burrows's Delta, text attribution, stylistic analysis, machine learningAbstract
The last two decades saw a dramatic increase in the number of papers published on the subject of stylometry, which is often narrowly understood as the task of identification of the author of a particular text fragment based on its stylistic properties. We present a new lightweight algorithm for stylometric identification of authors of Latin prose texts based on Burrows’s Delta, computed over relative frequencies of 244 manually selected genre and topic neutral words, and the Dirichlet distribution, whose parameters we estimate using an iterative maximum-likelihood algorithm. In order to demonstrate the effectiveness of the method, we present a case study of 3000-word fragments of texts by 36 classical and medieval authors and show that our method performs on par with Random Forest, a powerful general-purpose classification algorithm. We provide summary statistics of our algorithm’s performance together with confusion matrices demonstrating pairwise discriminability of texts by different authors. The advantages of our method are that it is very simple to implement, very quick to train and do inference with, and that it is very interpretable since it is a model-based algorithm: precision of the fitted Dirichlet distributions directly corresponds to the stylistic homogeneity of the texts by different authors. This makes it possible to use the algorithm as a general research tool in Latin stylistics.Downloads
References
Adams, J. N. (2013). Social variation and the Latin language. Cambridge Univ. Press.
Bela, A., Frigyik, A., & Gupta, M. (2010). Introduction to the Dirichlet distribution and related processes. Technical Report UWEETR-2010-0006. Department of Electrical Engineering, Univ. of Washington.
Blei, D. M., Ng, A. Y., & Jordan, M. I. (2003). Latent Dirichlet allocation. Journal of Machine Learning Research, 3(4-5), 993-1022.
Bolt, T. J., Flynt, J. H., Chaudhuri, P., & Dexter, J. P. (2019). A stylometry toolkit for Latin literature. In S. Padó, R. Huang (Eds.). Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP): System Demonstrations (pp. 205-210). Association for Computational Linguistics.
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5-32.
Burrows, J. (2002). ‘Delta': A measure of stylistic difference and a guide to likely authorship. Literary and Linguistic Computing, 17(3), 267-287.
Cameron, A. (2004). Greek mythography in the Roman world. Oxford Univ. Press.
Campbell, G., Corns, T. N., Hale, J. K., & Tweedie, F. J. (2007). Milton and the manuscript of De Doctrina Christiana. Oxford Univ. Press.
Chaudhuri, P., Dasgupta, T., Dexter, J. P., & Iyer, K. (2018). A small set of stylometric features differentiates Latin prose and verse. Digital Scholarship in the Humanities, 34(4), 716-729.
Cousin, J. (1943). Les langues spéciales. In Mémorial des études latins: Publié à l'occasion du vingtième anniversaire de la Société de la Revue des études latines, offert par la Société à son fondateur J. Marouzeau (pp. 37-54). Les belles lettres. (In French).
Ding, S. H. H., Fung, B. C. M., Iqbal, F., & Cheung, W. K. (2017). Learning stylometric representations for authorship analysis. IEEE Transactions on Cybernetics, 49(1), 107-21.
Eder, M., Rybicki, J., & Kestemont, M. (2016). Stylometry with R: A package for computational text analysis. The R Journal, 8(1), 107-121.
Gill, P. S., & Swartz, T. B. (2011). Stylometric analyses using Dirichlet process mixture models. Journal of Statistical Planning and Inference, 141(11), 3665-3674.
Gill, P. S., Swartz, T. B., & Treschow, M. (2007). A stylometric analysis of King Alfred's literary works. Journal of Applied Statistics, 34(10), 1251-1258.
Holmes, D. I. (1998). The evolution of stylometry in humanities scholarship. Literary and Linguistic Computing, 13(3), 111-117.
Holmes, D. I., & Kardos, J. (2003). Who was the author? An introduction to stylometry. Chance, 16(2), 5-8.
Kabala, J. (2020). Computational authorship attribution in medieval Latin corpora: The case of the Monk of Lido (ca. 1101-08) and Gallus Anonymous (ca. 1113-17). Language Resources and Evaluation, 54, 1-32.
Kestemont, M., Stover, J., Koppel, M., Karsdorp, F., & Daelemans, W. (2016). Authenticating the writings of Julius Caesar. Expert Systems with Applications, 63, 86-96.
Koppel, M., & Winter, Y. (2014). Determining if two documents are written by the same author. Journal of the Association for Information Science and Technology, 65(1), 178-187.
Kraus, M. A. (2017). Jewish, Christian, and classical exegetical traditions in Jerome's translation of the Book of Exodus: Translation technique and the vulgate. Brill.
Langslow, D. R. (2000). Medical Latin in the Roman Empire. Oxford Univ. Press.
Langslow, D. R. (2005). ‘Langues réduites au lexique?' The languages of Latin technical prose. In T. Reinhardt, M. Lapidge, J. N. Adams (Eds.). Aspects of the language of Latin prose
(pp. 287-302). Oxford Univ. Press.
Lebek, W. D. (1970). Verba prisca: Die Anfänge des Archaisierens in der lateinischen Beredsamkeit und Geschichtsschreibung. Vandenhoeck und Ruprecht. (In German).
Minka, T. P. (2012). Estimating a Dirichlet distribution. https://tminka.github.io/papers/dirichlet.
Mosteller, F., & Wallace, D. L. (1964). Applied Bayesian and classical inference: The case of the Federalist Papers. Addison-Wesley.
Rebenich, S. (2002). Jerome. Routledge.
Seroussi, Y., Zukerman, I., & Bohnert, F. (2014). Authorship attribution with topic models. Computational Linguistics, 40(2), 269-310.
Stover, J. A. (Ed. & Trans.) (2015). A new work by Apuleius: The lost third book of the De Platone. Oxford Univ. Press.
Stover, J. A., Winter, Y., Koppel, M., & Kestemont, M. (2016). Computational authorship verification method attributes a new work to a major 2nd century African author. Journal of the Association for Information Science and Technology, 67(1), 239-242.
Stover, J. A., & Kestemont, M. (2016a). Reassessing the Apuleian corpus: A computational approach to authenticity. The Classical Quarterly, 66(2), 645-72.
Stover, J. A., & Kestemont, M. (2016b). The authorship of the Historia Augusta: Two new computational studies. Bulletin of the Institute of Classical Studies, 59(2), 140-157.
Thilo, G. (Ed.). (1881-18877). Servii grammatici qui feruntur in Vergilii carmina commentarii. Teubner. (In Latin).
Vapnik, V. N. (1999). An overview of statistical learning theory. IEEE Transactions on Neural Networks, 10(5), 988-999.
Zetzel, J. E. G. (2018). Critics, compilers, and commentators: An introduction to Roman philology, 200 BCE - 800 CE. Oxford Univ. Press.
