Du, Keli ; Dudar, Julia ; Schöch, Christof (2023)
Evaluation of Measures of Distinctiveness. Classification of Literary Texts on the Basis of Distinctive Words.
In: Journal of Computational Literary Studies, 2022, 1 (1)
doi: 10.26083/tuprints-00023252
Artikel, Zweitveröffentlichung, Verlagsversion
Kurzbeschreibung (Abstract)
This paper concerns an empirical evaluation of nine different measures of distinctiveness or ‘keyness’ in the context of Computational Literary Studies. We use nine different sets of literary texts (specifically, novels) written in seven different languages as a basis for this evaluation. The evaluation is performed as a downstream classification task, where segments of the novels need to be classified by subgenre or period of first publication. The classifier receives different numbers of features identified using different measures of distinctiveness. The main contribution of our paper is that we can show that across a wide variety of parameters, but especially when only a small number of features is used, (more recent) dispersion-based measures very often outperform other (more established) frequency-based measures by significant margins. Our findings support an emerging trend to consider dispersion as an important property of words in addition to frequency.
| Typ des Eintrags: | Artikel |
|---|---|
| Erschienen: | 2023 |
| Autor(en): | Du, Keli ; Dudar, Julia ; Schöch, Christof |
| Art des Eintrags: | Zweitveröffentlichung |
| Titel: | Evaluation of Measures of Distinctiveness. Classification of Literary Texts on the Basis of Distinctive Words |
| Sprache: | Englisch |
| Publikationsjahr: | 2023 |
| Ort: | Darmstadt |
| Publikationsdatum der Erstveröffentlichung: | 2022 |
| Verlag: | Universitäts- und Landesbibliothek Darmstadt |
| Titel der Zeitschrift, Zeitung oder Schriftenreihe: | Journal of Computational Literary Studies |
| Jahrgang/Volume einer Zeitschrift: | 1 |
| (Heft-)Nummer: | 1 |
| Kollation: | 21 Seiten |
| DOI: | 10.26083/tuprints-00023252 |
| URL / URN: | https://tuprints.ulb.tu-darmstadt.de/23252 |
| Zugehörige Links: | |
| Herkunft: | Zweitveröffentlichung von TUjournals |
| Kurzbeschreibung (Abstract): | This paper concerns an empirical evaluation of nine different measures of distinctiveness or ‘keyness’ in the context of Computational Literary Studies. We use nine different sets of literary texts (specifically, novels) written in seven different languages as a basis for this evaluation. The evaluation is performed as a downstream classification task, where segments of the novels need to be classified by subgenre or period of first publication. The classifier receives different numbers of features identified using different measures of distinctiveness. The main contribution of our paper is that we can show that across a wide variety of parameters, but especially when only a small number of features is used, (more recent) dispersion-based measures very often outperform other (more established) frequency-based measures by significant margins. Our findings support an emerging trend to consider dispersion as an important property of words in addition to frequency. |
| Freie Schlagworte: | keyness, evaluation, literary texts, distinctiveness |
| Status: | Verlagsversion |
| URN: | urn:nbn:de:tuda-tuprints-232529 |
| Zusätzliche Informationen: | Urspr. Konferenzveröffentlichung/Originally conference publication: 1st Annual Conference of Computational Literary Studies, 01.-02.06.2022, Darmstadt, Germany |
| Sachgruppe der Dewey Dezimalklassifikatin (DDC): | 800 Literatur > 800 Literatur, Rhetorik, Literaturwissenschaft |
| Fachbereich(e)/-gebiet(e): | 02 Fachbereich Gesellschafts- und Geschichtswissenschaften > Institut für Sprach- und Literaturwissenschaft > Digital Philology - Neuere deutsche Literaturwissenschaft 02 Fachbereich Gesellschafts- und Geschichtswissenschaften 02 Fachbereich Gesellschafts- und Geschichtswissenschaften > Institut für Sprach- und Literaturwissenschaft |
| Hinterlegungsdatum: | 21 Feb 2023 10:19 |
| Letzte Änderung: | 27 Feb 2023 12:46 |
| PPN: | |
| Zugehörige Links: | |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum