Do There Exist an Emotion Trend in Scientific Papers? PRO-VE Conference as a Case
Authors: Venumuddala, R., Xu, L., de Vrieze, P.
Journal: IFIP Advances in Information and Communication Technology
Publication Date: 01/01/2024
Volume: 726 IFIP
Pages: 118-133
eISSN: 1868-422X
ISSN: 1868-4238
DOI: 10.1007/978-3-031-71739-0_8
Abstract:Scientific writing aims for formality and objectivity, yet emotions are integral to human communication, decision-making, and collaboration, all of which are fundamental to scientific progress. Existing research on emotion detection has mainly focused on datasets from social media and online platforms, where emotional expressions are abundant. However, scientific texts pose unique challenges due to their formal language and the rarity of explicit emotional words, necessitating specialised investigation. This study investigates the presence and nature of emotions in scientific texts, specifically analysing the abstracts from the PRO-VE conference series from 2012 to 2022. Two emotion detection methods are employed: a lexicon-based approach and a hybrid machine learning-based approach. The lexicon-based approach utilises the NRC Emotion Lexicon to identify and quantify emotions within the PRO-VE abstracts, while the hybrid approach integrates Word2Vec for word embedding generation and a Random Forest classifier for emotion prediction. The findings reveal a predominance of positive emotions, such as trust, anticipation, and joy, in the PRO-VE abstracts, consistent with the objective nature of scientific writing. In light of the PRO-VE conference series’ 25th anniversary, an analysis of trends and patterns in the detected emotions offers insights into the emotional landscape of this prestigious conference series. The study also critically examines the limitations of the experiments, including the dataset size and the prevalence of positive emotions.
https://eprints.bournemouth.ac.uk/40332/
Source: Scopus
Do There Exist an Emotion Trend in Scientific Papers? PRO-VE Conference as a Case
Authors: Venumuddala, R., Xu, L., de Vrieze, P.
Journal: NAVIGATING UNPREDICTABILITY: COLLABORATIVE NETWORKS IN NON-LINEAR WORLDS, PRO-VE 2024, PT I
Publication Date: 2024
Volume: 726
Pages: 118-133
eISSN: 1868-422X
ISBN: 978-3-031-71741-3
ISSN: 1868-4238
DOI: 10.1007/978-3-031-71739-0_8
https://eprints.bournemouth.ac.uk/40332/
Source: Web of Science
Do There Exist an Emotion Trend in Scientific Papers? PRO-VE Conference as a Case
Authors: Xu, L., de Vrieze, P.
Conference: 25th FIFP/SOCOLNET working Conference on Virtual Enterprises (PRO-VE 2024)
Dates: 28/10/2024
Publication Date: 01/10/2024
Publisher: Springer
Abstract:Scientific writing aims for formality and objectivity, yet emotions are integral to human communication, decision-making, and collaboration, all of which are fundamental to scientific progress. Existing research on emotion detection has mainly focused on datasets from social media and online platforms, where emotional expressions are abundant. However, scientific texts pose unique challenges due to their formal language and the rarity of explicit emotional words, necessitating specialised investigation. This study investigates the presence and nature of emotions in scientific texts, specifically analysing the abstracts from the PRO-VE conference series from 2012 to 2022. Two emotion detection methods are employed: a lexicon-based approach and a hybrid machine learning-based approach. The lexicon-based approach utilises the NRC Emotion Lexicon to identify and quantify emotions within the PRO-VE abstracts, while the hybrid approach integrates Word2Vec for word embedding generation and a Random Forest classifier for emotion prediction. The findings reveal a predominance of positive emotions, such as trust, anticipation, and joy, in the PRO-VE abstracts, consistent with the objective nature of scientific writing. In light of the PRO-VE conference series' 25th anniversary, an analysis of trends and patterns in the detected emotions offers insights into the emotional landscape of this prestigious conference series. The study also critically examines the limitations of the experiments, including the dataset size and the prevalence of positive emotions.
https://eprints.bournemouth.ac.uk/40332/
Source: Manual