Automatic literary school assignment
Linguistic-statistical studies of lusophone literature
Abstract
In this paper we use a set of syntactic and semantic features of Portuguese to automatically classify literary works in literary periods and/or schools, and address the issue of their appropriateness, for two different literary collections.
The first task attempts to replicate the work by Barufaldi and colleagues, who applied compression methods on 37 Brazilian works by 15 different authors and classified the works in 4 different literary schools.
The second collection, of 192 novels published in Portugal and Brazil in the period 1840 to 1919, features many works who cannot be singly accomodated in one literary school only, and which have been (not mutually exclusively) classified as romantic, realist, naturalist, symbolist, decadent and modernist.
We use classification techniques in R, such as discriminant analysis and support vector models for the first task, and correspondence analysis for the second collection. We also apply topic modeling to (distinct subsets of) the second collection in order to investigate whether this technique can provide us with recurrent topics for different literary schools.
Copyright (c) 2020 Diana Santos, Emanoel Pires, Cláudia Freitas, Rebeca Schumacher Fuão, João Marques Lopes
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
