Widok zawartości stron Widok zawartości stron

Widok zawartości stron Widok zawartości stron

Authorship attribution and authorship verification in Indology: traditional philology and computational stylometry

Termin: 27.05.2019
Miejsce: Kolegium Śląskie, al. Mickiewicza 3, s. 2 (parter)
Organizator: Zakład Języków i Kultur Indii i Azji Południowej

Zakład Języków i Kultur Indii i Azji Południowej

zaprasza na wykład:

 

 

Prof. Ivan Andrijanić (University of Zagreb)
 
 

Authorship attribution and authorship verification in Indology: traditional philology and computational stylometry

 

 

The term stylometry, as introduced by Polish philosopher and philologist Wincenty Lutosławski in 1890, implies the measurement of textual stylistic affinities in order to address different literary issues such as authorship, chronology, or text-history. In authorship studies, the stylistic characteristics of a text of unknown authorship are compared to the determined profiles of known authors. This type of categorisation is usually referred to as authorship attribution in authorship studies. In the past 30 years, computational stylometrics has seen significant progress and numerous techniques have been proposed, culminating in machine-learning classification models. In authorship attribution studies, authorial profiles are selected and compared to the profile of a text by an unknown or disputed author in order to identify a matching candidate.

This lecture will briefly present the idea and the history of stylometric studies, with the focus on possible implementation of existing stylometric statistical techniques on Sanskrit texts of unknown or disputed authorship. Lately, an ever larger body of Sanskrit electronic texts (GRETIL, TITUS etc.) have emerged together with the development of reliable authorship verification algorithms. However, unlike other languages, Sanskrit offers some serious challenges that will be presented in the lecture. At the end, the results of experiments with Sanskrit texts conducted in the stylo package (Eder et al. 2016), an open source stylometric script written in the statistical programming environment R, will be presented.

 

 

27.05.2019 (poniedziałek), godz. 13.00

Data opublikowania: 02.05.2019
Osoba publikująca: Anna Nitecka