Sentiment Analysis of Serbian Old Novels

Објеката

Тип
Рад у зборнику
Верзија рада
објављена
Језик
енглески
Креатор
Ranka Stanković, Miloš Košprdić, Milica Ikonić Nešić, Tijana Radović
Извор
Proceedings of the 2nd Workshop on Sentiment Analysis and Linguistic Linked Data, June 2022, Marseille, France
Издавач
European Language Resources Association
Датум издавања
2022
Сажетак
In this paper we present first study of Sentiment Analysis (SA) of Serbian novels from the 1840-1920 period. The preparation of sentiment lexicon was based on three existing lexicons: NRC, AFFIN and Bing with additional extensive corrections. The first phase of dataset refinement included filtering the word that are not found in Serbian morphological dictionary and in second automatic POS tagging and lemma were manually corrected. The polarity lexicon was extracted and transformed into ontolex-lemon and published as initial version. The complex inflection system of Serbian language required expansion of sentiment lexicon with inflected forms from Serbian morphological dictionaries. Set of sentences for SA was extracted from 120 novels of Serbian part of ELTeC collection, labelled for polarity and used for several model training. Several approaches for SA are compared, starting with for variation of lexicon based and followed by Logistic Regression, Naive Bayes, Decision Tree, Random Forest, SVN and k-NN. The comparison with models trained on labelled movie reviews dataset indicates that it can not successfully be used for sentiment analysis of sentences in old novels.
почетак странице
31
крај странице
38
Subject
sentiment lexicon, sentiment analysis, distant-reading, machine learning, old novels
Шира категорија рада
М30
Ужа категорија рада
М33
Права
Отворени приступ
Лиценца
Creative Commons – Attribution-NonComercial-Share Alike 4.0 International
Формат
.pdf
Скупови објеката
Ранка Станковић
Radovi istraživača
Медија
2022.salld2-1.6.pdf

Ranka Stanković, Miloš Košprdić, Milica Ikonić Nešić, Tijana Radović. "Sentiment Analysis of Serbian Old Novels" in Proceedings of the 2nd Workshop on Sentiment Analysis and Linguistic Linked Data, June 2022, Marseille, France, European Language Resources Association (2022)

This item was submitted on 8. јул 2022. by [anonymous user] using the form “Рад у зборнику радова” on the site “Радови”: http://drug.rgf.bg.ac.rs/s/repo

Click here to view the collected data.