Harmini, Triana (2020) Sundanese Twitter Dataset for Emotion Classification. Sundanese Twitter Dataset for Emotion Classification. ISSN 978-1-7281-8283-4

	FILE TEXT (jurnal) 10. Sundanese Twitter Dataset.pdf - Published Version License Creative Commons Attribution Non-commercial Share Alike. Download (518kB)
	FILE TEXT (riviewer) 10. sundanese twitter dataset.pdf Download (1MB)
	FILE TEXT (plagiarism) 10. Sundanese Twitter Dataset.pdf Download (1MB)

Abstract

Sundanese is the second-largest tribe in Indonesia which possesses many dialects. This condition has gained attention for many researchers to analyze emotion especially on social media. However, with barely available Sundanese dataset, this condition makes understanding sundanese emotion is a challenging task. In this research, we proposed a dataset for emotion classification of Sundanese text. The preprocessing includes case folding, stopwords removal, stemming, tokenizing, and text representation. Prior to classification, for the feature generation, we utilize term frequency-inverse document frequency (TFIDF). We evaluated our dataset using k-Fold Cross Validation. Our experiments with the proposed method exhibit an effective result for machine learning classification. Furthermore, as far as we know, this is the first Sundanese emotion dataset available forpublic

View more statistics

View Item