RoEmoLex - A Romanian Emotion Lexicon

Briciu, Anamaria; Lupea, Mihaiela

doi:10.24193/subbi.2017.2.04

Cited by 9 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Therefore, most of the available data and tools are for English. There are various studies conducted on languages such as Chinese [18], Arabic [19], French [20] and Romanian [21].…”

Section: Related Studiesmentioning

confidence: 99%

A study of Turkish emotion classification with pretrained language models

Uçan

Dörterler

Sezer

2021

Journal of Information Science

View full text Add to dashboard Cite

Emotion classification is a research field that aims to detect the emotions in a text using machine learning methods. In traditional machine learning (TML) methods, feature engineering processes cause the loss of some meaningful information, and classification performance is negatively affected. In addition, the success of modelling using deep learning (DL) approaches depends on the sample size. More samples are needed for Turkish due to the unique characteristics of the language. However, emotion classification data sets in Turkish are quite limited. In this study, the pretrained language model approach was used to create a stronger emotion classification model for Turkish. Well-known pretrained language models were fine-tuned for this purpose. The performances of these fine-tuned models for Turkish emotion classification were comprehensively compared with the performances of TML and DL methods in experimental studies. The proposed approach provides state-of-the-art performance for Turkish emotion classification.

show abstract

“…Therefore, most of the available data and tools are for English. There are various studies conducted on languages such as Chinese [18], Arabic [19], French [20] and Romanian [21].…”

Section: Related Studiesmentioning

confidence: 99%

A study of Turkish emotion classification with pretrained language models

Uçan

Dörterler

Sezer

2021

Journal of Information Science

View full text Add to dashboard Cite

show abstract

“…Another approach is RoEmoLex (Briciu & Lupea, 2017) that uses the Romanian version of the English words from EmoLex (the words are automatically translated) with a series of additional processing to clean the data (deleting words without valences, deleting duplicates, adding tags, parts of speech, translating missing words / manual validation of the translation etc.). In this work, RoWordNet was also employed in order to assign the corresponding synset along with other resources (SentiWordNet scores (Esuli & Fabrizio, 2006), SUMO categories etc.)…”

Section: Literature Reviewmentioning

confidence: 99%

Towards the Development of a Romanian Lexicon for the Analysis of Emotions in the Literary Works of Canonical Authors

Gavrilă¹,

Băjenaru²,

Dobre³

et al. 2021

STUD INFORM CONTROL

View full text Add to dashboard Cite

This paper focuses on extracting emotions from works published by Romanian canonical authors. The information about canonical authors represents a part of the data provided by the INTELLIT platform (Gavrilă, Băjenaru & Dobre, 2019a). The INTELLIT platform aims to be the largest digital archive in Romania containing data regarding our national literary heritage. All the information is provided by the Romanian Academy, the Institute of Literary History and Theory and the National Foundation for Science and Art in collaboration with the Publishing House of the Museum of Romanian Literature (Simion, 2005; Simion, 2006). The platform provides information about the life and work of the Romanian authors, about important events in our culture (complete chronology of events from 1994 to 2000), information about all the authors from A to Z, different works written

show abstract

“…With the emotions being determined and the expression of solidarity and ostracism through emotions being defined, it now needs to be elucidated which words can be associated with the emotions. Therefore, the analysis avails the Romanian Emotion Lexicon (RoEmoLex) Version 3 that includes almost 10,000 Romanian words and idioms which are related to positivity, negativity as well as to the eight basic emotions of Plutchik's wheel of emotions (Briciu and Lupea 2017;2019).…”

Section: Emotions In Political Language Usementioning

confidence: 99%

Emotions in the political discourse of Romania. A corpus-driven analysis of multiword expressions

Bíro

2020

BUT PCS

View full text Add to dashboard Cite

Emotions in political language use can have considerable repercussions on a society, especially on the expression of solidarity and ostracism, which are closely connected to the emotions of trust and aversion. Driven2 by a sub-corpus of spoken presidential texts from 1992 to 2004, this analysis constructs on the basis of multiword expressions emotion profiles of the political actors. To do so, it draws from Robert Plutchik’s wheel of emotions3 and the Romanian Emotion Lexicon (RoEmoLex)4 and focuses on hints for trust and aversion. The findings indicate a high degree of expressions of trust while aversion is nearly non-existent. Moreover, an increase of emotions during the three presidential terms of Ion Iliescu and Emil Constantinescu is remarkable from a diachronic perspective and constitutes a transition from the topics of internal stability and security to global cooperation and responsibility.

show abstract

RoEmoLex - A Romanian Emotion Lexicon

Cited by 9 publications

References 7 publications

A study of Turkish emotion classification with pretrained language models

A study of Turkish emotion classification with pretrained language models

Towards the Development of a Romanian Lexicon for the Analysis of Emotions in the Literary Works of Canonical Authors

Emotions in the political discourse of Romania. A corpus-driven analysis of multiword expressions

Contact Info

Product

Resources

About