Parallel Stylometric Document Embeddings with Deep Learning Based Language Models in Literary Authorship Attribution

Објеката

Тип
Рад у часопису
Верзија рада
објављена верзија
Језик
енглески
Креатор
Mihailo Škorić, Ranka Stanković, Milica Ikonić Nešić, Joanna Byszuk, Maciej Eder
Извор
Mathematics
Издавач
MDPI AG
Датум издавања
2022
том
10
издање
5
doi
10.3390/math10050838
issn
2227-7390
Subject
General Mathematics, Engineering (miscellaneous), Computer Science (miscellaneous)
Шира категорија рада
M20
Ужа категорија рада
М21а
Права
Отворени приступ
Лиценца
All rights reserved
Формат
.pdf
Сажетак
This paper explores the effectiveness of parallel stylometric document embeddings in solving the authorship attribution task by testing a novel approach on literary texts in 7 different languages, totaling in 7051 unique 10,000-token chunks from 700 PoS and lemma annotated documents. We used these documents to produce four document embedding models using Stylo R package (word-based, lemma-based, PoS-trigrams-based, and PoS-mask-based) and one document embedding model using mBERT for each of the seven languages. We created further derivations of these embeddings in the form of average, product, minimum, maximum, and l2 norm of these document embedding matrices and tested them both including and excluding the mBERT-based document embeddings for each language. Finally, we trained several perceptrons on the portions of the dataset in order to procure adequate weights for a weighted combination approach. We tested standalone (two baselines) and composite embeddings for classification accuracy, precision, recall, weighted-average, and macro-averaged F1-score, compared them with one another and have found that for each language most of our composition methods outperform the baselines (with a couple of methods outperforming all baselines for all languages), with or without mBERT inputs, which are found to have no significant positive impact on the results of our methods.

Mihailo Škorić, Ranka Stanković, Milica Ikonić Nešić, Joanna Byszuk, Maciej Eder. "Parallel Stylometric Document Embeddings with Deep Learning Based Language Models in Literary Authorship Attribution" in Mathematics, MDPI AG (2022). https://doi.org/10.3390/math10050838

This item was submitted on 7. март 2022. by [anonymous user] using the form “Рад у часопису” on the site “Радови”: http://romeka.rgf.rs/s/repo

Click here to view the collected data.