Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal Jurnal Nasional Komputasi dan Teknologi Informasi Jurnal Netnografi Komunikasi
Annisa Damayanti
Universitas Dian Nusantara
Aerospace Engineering Arts Humanities Automotive Engineering Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Engineering Languange, Linguistic, Communication & Media Library & Information Science Neuroscience Social Sciences Transportation

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Integrasi Model RoBERTa untuk Analisis Sentimen dan Deteksi Ujaran Kebencian pada Komentar YouTube: Pendekatan NLP dan Netiket Syukriyansyah Syukriyansyah; Annisa Damayanti
Jurnal Nasional Komputasi dan Teknologi Informasi (JNKTI) Vol 8, No 4 (2025): Agustus 2025
Publisher : Program Studi Teknik Komputer, Fakultas Teknik. Universitas Serambi Mekkah

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32672/jnkti.v8i4.9718

Abstract

Abstrak − Isu lingkungan seperti pertambangan di Papua memicu diskusi luas di YouTube, namun sering diwarnai ujaran kebencian yang melanggar prinsip etika digital (netiket). Penelitian ini menganalisis sentimen dan mendeteksi ujaran kebencian dalam komentar YouTube menggunakan pendekatan Natural Language Processing (NLP). Sebanyak 3.827 komentar dari kanal Greenpeace Indonesia dikumpulkan dan dipraproses melalui tahapan konversi huruf kecil, penghapusan tag HTML dan URL, penghapusan angka, normalisasi kata tidak baku, penghapusan simbol non-alfabetik, stemming dengan Sastrawi Stemmer, dan penghapusan stopword. Analisis sentimen dilakukan dengan model w11wo/indonesian-roberta-base-sentiment-classifier, sedangkan deteksi ujaran kebencian menggunakan model OwLim/indonesian-roberta-hate-speech. Hasil menunjukkan 57,7% komentar bersentimen negatif, dan 52,4% di antaranya mengandung ujaran kebencian. Analisis linguistik mengungkap strategi dehumanisasi, kosakata ofensif, dan tuduhan kolektif yang melanggar netiket. Penelitian ini menyoroti perlunya integrasi teknologi moderasi konten berbasis NLP dengan pendidikan etika digital untuk menciptakan ekosistem diskusi yang sehat dan beradab di ruang digital.Kata Kunci: Ujaran Kebencian; NLP; Analisis Sentimen; YouTube; Netiket; Etika Digital; Abstract - Environmental issues such as mining in Papua spark extensive discussions on YouTube, but are often marred by hate speech violating digital ethics principles (netiquette). This study analyzes sentiment and detects hate speech in YouTube comments using a Natural Language Processing (NLP) approach. A total of 3,827 comments from Greenpeace Indonesia's channel were collected and preprocessed through lowercase conversion, HTML and URL removal, number removal, informal word normalization, non-alphabetic symbol removal, stemming using Sastrawi Stemmer, and stopword removal. Sentiment analysis was conducted using the w11wo/indonesian-roberta-base-sentiment-classifier model, while hate speech detection employed the OwLim/indonesian-roberta-hate-speech model. Results show 57.7% of comments had negative sentiment, with 52.4% containing hate speech. Linguistic analysis revealed dehumanization strategies, offensive vocabulary, and collective accusations that violate netiquette. This research highlights the need to integrate NLP-based content moderation technology with digital ethics education to create a healthy and civilized discussion ecosystem in digital spaces. Keywords: Hate Speech, NLP, Sentiment Analysis, YouTube, Netiquette, Digital Ethics;
MORAL REASONING AND DISCOURSE FRAGMENTATION IN YOUTUBE COMMENTS ON SEXUAL VIOLENCE CASES INVOLVING RELIGIOUS AUTHORITY Damayanti, Annisa; Syukriyansyah
Jurnal Netnografi Komunikasi Vol. 4 No. 2 (2025): Vol. 4 No. 2 (2025): JNK National Accredited Rank. SINTA 5 based on SK Kemdikti
Publisher : Communication Science Department - Faculty of Social and Political Sciences, Universitas Satya Negara Indonesia (USNI)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59408/jnk.v4i2.124

Abstract

This study examines moral reasoning and discourse fragmentation within the YouTube comment section of a viral video detailing a case of sexual violence involving a religious authority figure in Indonesia. Employing a Mixed Methods Social Network Analysis (MMSNA) approach, we analyzed 6,018 comments from 5,484 users to map the network structure, identify communicative communities, and assess emotional polarization via sentiment analysis. Findings reveal a highly fragmented and sparse network with low user interactivity, dominated by isolated, small-scale communities. Despite this structural fragmentation, sentiment analysis showed a predominance of neutral expressions (76.9%), with limited emotional polarization between the six main thematic communities identified. These communities function as distinct "affective micro-publics," articulating responses through specific discursive roles: moral-religious condemnation, emotional support, social reflection, fear of stigma, digital activism, and social ethics critique. The study concludes that the digital discourse operates not as a unified deliberative space but as a constellation of value-aligned clusters, where morality and empathy, rather than rational debate, mediate public participation. This underscores the role of platform architecture in fostering affective enclaves around sensitive socio-religious issues.
Co-Authors Syukriyansyah Syukriyansyah Syukriyansyah