Correlations and Potential Cross-Linguistic Indicators of Writing Style

Juola, Patrick; Mikros, George K.; Vinsick, Sean

doi:10.1080/09296174.2018.1458395

Cited by 23 publications

(6 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although AMNP and WE features are very successful in the monolingual datasets, their classification accuracy in the cross-linguistic experiment was the same as the baseline, meaning that they did not perform better than selecting author labels purely by chance. This result confirms previous research in cross-linguistic authorship attribution (Juola, Mikros, & Vinsick, 2019a, 2019b, in which patterns of a strong correlation between Greek, Spanish and English QL indices have been observed.…”

Section: Authorship Attribution Experimentssupporting

confidence: 92%

Cross-linguistic Authorship Attribution and Author Profiling. Is Machine Translation the Solution?

Mikros

Boumparis

2022

Preprint

View full text Add to dashboard Cite

This paper aims to explore whether cross-linguistic authorship attribution and author's gender identification are feasible using Machine Translation (MT) as a method to bridge the language gap. We designed a series of computational stylistics experiments to explore whether the stylometric signal survives through the MT process. We compiled an extensive blog corpus in Greek containing 100 authors, balanced in gender with 50 texts from each author. Then, we used Google's Neural Machine Translation to automatically translate each text into English. We ran several classification experiments using the Random Forest algorithm in authorship attribution and gender profiling tasks employing different feature groups in both the source language (Greek) and the machine-translated (English) corpora. Moreover, we trained models in the source language and used the texts in the target language as the unseen test set, to simulate a cross-linguistic prediction. The results showed that cross-linguistic gender identification could be achieved reliably using n-gram features. Authorship attribution, however, was not feasible in our cross-linguistic experimental setting, although features of lexical diversity exhibited promising results. Finally, when performed using training and testing set from the same language, both authorship attribution and gender identification exhibit similar accuracy across the full range of the feature groups used.

show abstract

Section: Authorship Attribution Experimentssupporting

confidence: 92%

Cross-linguistic Authorship Attribution and Author Profiling. Is Machine Translation the Solution?

Mikros

Boumparis

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Various studies have proposed to describe the style of each author by a few vocabulary richness indicators (Juola et al, 2019) such as Type-Token Ratio (TTR) (Hart, 1984), hapax density (percentage of words occurring once), or the lexical density (LD) (Biber et al, 2002) defined in Equation 1.…”

Section: State Of the Artmentioning

confidence: 99%

Stylistic analysis of the French presidential speeches: Is Macron really different?

Labbé

Savoy

2020

Digital Scholarship in the Humanities

View full text Add to dashboard Cite

Presidential speeches indicate the government’s intentions and justifications supported by a dedicated style and rhetoric oscillating between explanation and controversy. Over a period of 60 years, can we observe stylistic variations by the different French presidents of the Fifth Republic (1958–2018)? Based on official transcripts of all their allocutions, this article illustrates the stylistic evolution and presents the underlying main trends. This study shows that de Gaulle’s rhetoric is not mainly dedicated to his own person, or that the two terms of J. Chirac are not fully similar. According to several overall stylistic indicators, Macron’s style does not appear as complex compared to his predecessors (F. Hollande or N. Sarkozy) but a more careful analysis clearly demonstrates his noticeable new style. Compared to the recent US presidents, the French ones present some similarities (e.g. similar mean sentence length) and dissimilarities (more I-words, less we-words). In this comparative analysis, Macron’s style is also clearly distinctive from both the US and former French presidents. Opting for a more abstract discourse, less anchored in space, using less numbers, E. Macron tends to use long sentences. These various stylistic and rhetorical features could explain his being misunderstood by the French people and his recurrent low approval ratings.

show abstract

“…Classification of texts with respect to author's identity, author's sex, genre, time of writing, certain language features (e. g., analyticity/syntheticity), etc. remains a topical task despite various levels of success achieved upon application of efforts so far (Tuldava 2004;Kelih et al 2005;Koppel et al 2009;Nini and Grant 2013;Zörnig et al 2016;Juola et al 2018;Shen and Tao 2021). Potential results are promising both in purely theoretical and practical prospect, for example in machine learning, sociolinguistics, literary studies, forensic applications and so on.…”

Section: Introductionmentioning

confidence: 99%

“…Our study is based on the experience of several researcher groups as described below. An analysis of a Twitter corpus and a corpus of essays was made by Juola et al (2018). A stylometric analysis of British journalistic texts by Cortina-Borja and Chappas (2006) was based on the distribution of nouns only.…”

Section: Introductionmentioning

confidence: 99%

Attempting at parametrization of moderate-length poetic texts: Moses, a poem by Ivan Franko

Buk¹,

Rovenchak²

2022

Glottometrics

View full text Add to dashboard Cite

The aim of this study is to find parameters that can be used for classification of not very long texts, for example, by author, genre, etc. We go through various known parameters and analyze to what extent they are useful for the intended purposes. We also suggest some improvements that need to be checked further. We calculate the values of parameters at various points of text comprising N tokens (running words) counted from the beginning of text. As parameters with prospects of author and/or language attribution we identify, in particular, the h-point scaling coefficient, Yule’s K, relative repeat rate, and the fraction of dis legomena. These parameters demonstrate quite stable behavior in N. Another set includes scaling exponents of parameters with respect to N. Certain modifications are suggested for Lambda and entropy introducing logarithmic corrections being powers of ln N. The results are applicable for texts of thousands to tens of thousand words.

show abstract

Correlations and Potential Cross-Linguistic Indicators of Writing Style

Cited by 23 publications

References 24 publications

Cross-linguistic Authorship Attribution and Author Profiling. Is Machine Translation the Solution?

Cross-linguistic Authorship Attribution and Author Profiling. Is Machine Translation the Solution?

Stylistic analysis of the French presidential speeches: Is Macron really different?

Attempting at parametrization of moderate-length poetic texts: Moses, a poem by Ivan Franko

Contact Info

Product

Resources

About