Stylometric Evaluation of Parameters and Distance Measures for Hispanic Texts

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F26%3AE7KQL2RS" target="_blank" >RIV/00216208:11320/26:E7KQL2RS - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.4324/9781003393771-10" target="_blank" >http://dx.doi.org/10.4324/9781003393771-10</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.4324/9781003393771-10" target="_blank" >10.4324/9781003393771-10</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Stylometric Evaluation of Parameters and Distance Measures for Hispanic Texts

Popis výsledku v původním jazyce

The selection of parameters and distance measures is one of the most important steps in conducting a stylometric research. Previous studies on the evaluation of these for different literary traditions and genres have found that their effectiveness varies depending on the language and genre of the texts. Although most of these studies have been carried out for European languages, Hispanic texts are rarely addressed. Our study aims to fill this important gap in Hispanic texts, and to provide Hispanic literature researchers with the best performing parameters for stylometry and authorship attribution studies. To do so, we carry out an evaluation of the distance measures that have been used the most and obtained best results in general cross-linguistic research. We also evaluate different parameters such as the number of most frequent words (from 100 to 5,000), the kind of n-grams (unigrams, bigrams and trigrams), and if linguistic annotation of certain features (such as part-of-speech) improves the results of authorship attribution. The evaluation is conducted through supervised authorship attribution methods as implemented in the stylo package. © 2025 selection and editorial matter, Roberto J. González Zalacain and Gael Vaamonde; individual chapters, the contributors. All rights reserved.

Název v anglickém jazyce

Stylometric Evaluation of Parameters and Distance Measures for Hispanic Texts

Popis výsledku anglicky

The selection of parameters and distance measures is one of the most important steps in conducting a stylometric research. Previous studies on the evaluation of these for different literary traditions and genres have found that their effectiveness varies depending on the language and genre of the texts. Although most of these studies have been carried out for European languages, Hispanic texts are rarely addressed. Our study aims to fill this important gap in Hispanic texts, and to provide Hispanic literature researchers with the best performing parameters for stylometry and authorship attribution studies. To do so, we carry out an evaluation of the distance measures that have been used the most and obtained best results in general cross-linguistic research. We also evaluate different parameters such as the number of most frequent words (from 100 to 5,000), the kind of n-grams (unigrams, bigrams and trigrams), and if linguistic annotation of certain features (such as part-of-speech) improves the results of authorship attribution. The evaluation is conducted through supervised authorship attribution methods as implemented in the stylo package. © 2025 selection and editorial matter, Roberto J. González Zalacain and Gael Vaamonde; individual chapters, the contributors. All rights reserved.

Druh

C - Kapitola v odborné knize

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

—

Rok uplatnění

2025

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název knihy nebo sborníku

Digital Humanities in Medieval and Early Modern Spanish Texts: Current Perspectives and Approaches

ISBN

978-1-040-32811-8

Počet stran výsledku

26

Strana od-do

163-188

Počet stran knihy

300

Název nakladatele

Taylor and Francis

Místo vydání

—

Kód UT WoS kapitoly

—