Journals / Konya Mühendislik Bilimleri Dergisi / 2022 / Cilt: 10 - Sayı: 4

Aggression Detection in Twitter Data Using Transformer‑Based Convolutional Neural Network Model

TRANSFORMATÖR-TABANLI EVRİŞİMLİ SİNİR AĞI MODELİ KULLANARAK TWITTER VERİSİNDE SALDIRGANLIK TESPİTİ

Pages
986–1001
Publication Date
2022-12-03
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The performance parameters obtained in the experimental results were determined as accuracy, precision, recall, F1-score, and AUC, and it was revealed that the LMTweets + CNN model performed better among all the models used.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Online environments facilitate the increase of anti-social behaviors in people's social interactions.
📊
Key Findings & Data:Behaviors such as hate speech, cyberbullying, and trolling have increased significantly, especially in recent years, with the widespread use of social media.
💡
Conclusion & Impact:The performance parameters obtained in the experimental results were determined as accuracy, precision, recall, F1-score, and AUC, and it was revealed that the LMTweets + CNN model performed better among all the models used.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Saldırganlık · CNN · Derin öğrenme

Abstract

Online environments facilitate the increase of anti-social behaviors in people's social interactions. Behaviors such as hate speech, cyberbullying, and trolling have increased significantly, especially in recent years, with the widespread use of social media. Detection of aggression and hateful speech is an important step in reducing and preventing cyberbullying. Cyberbullying is defined as comments made on social media to harm other individuals by using hateful, offensive, rude, humiliating, and sarcastic expressions. It is slow and expensive to try to achieve this with human control with the existence of rapidly growing data, so cyberbullying can be stopped by automatic detection of aggression. In this study, the automatic determination of aggression detection via Cyber-Trolls, which is the Twitter dataset, is discussed. A coder named LMTweets was trained on 20001 tweets to extract the features of the dataset. The extracted features are given as input to the convolutional neural network model to classify the text as aggressive / non-aggressive. In addition, three classification algorithms, namely Naïve Bayes, Support Vector Machine, K-Nearest Neighbors, were applied. In addition, three transformer models, BERT, XLNet, and ULMFIT were applied along with the Convolutional Neural Network, Long Short-Term Memory, and Gated Recurrent Unit three learning algorithms. Python, Keras API, and Tensorflow are used together in the proposed model. The performance parameters obtained in the experimental results were determined as accuracy, precision, recall, F1-score, and AUC, and it was revealed that the LMTweets + CNN model performed better among all the models used.

Özet

Çevrimiçi ortamlar, insanların sosyal etkileşimlerinde anti-sosyal davranışların artmasını kolaylaştırmaktadır. Sosyal medya kullanımının yaygınlaşmasıyla özellikle son yıllarda nefret söylemleri, siber zorbalık ve trolleme gibi davranışlar önemli ölçüde artmıştır. Saldırgan ve nefret içerikli söylemlerin tespiti siber zorbalıkların azaltılması ve engellenmesinde önemli bir adımdır. Siber zorbalık, sosyal medya üzerinden nefret dolu, saldırgan, kaba, aşağılayıcı ve alaycı ifadeler kullanarak diğer bireylere zarar vermek adına yapılan yorumlar olarak adlandırılmaktadır. Hızla büyüyen verilerin varlığı, bunun insan denetimiyle gerçekleştirilmeye çalışılması yavaş ve pahalı olduğundan saldırganlığın otomatik tespitiyle siber zorbalığın durdurulması sağlanabilir. Bu çalışmada Twitter veri seti olan Cyber-Trolls üzerinden saldırganlık tespitini otomatik olarak belirlenmesi ele alınmaktadır. LMTweets adında bir kodlayıcı, veri kümesinin özelliklerinin çıkarılması için 20001 adet tweet üzerinden eğitilmiştir. Çıkarılan öznitelikler, metni saldırgan / saldırgan olmayan olarak sınıflandırmak üzere evrişim sinir ağı modeline girdi olarak verilir. Ayrıca Naïve Bayes, Destek Vektör Makinesi, K-En Yakın Komşu, olmak üzere üç sınıflandırma algoritması uygulanmıştır. Bunun yanında, Evrişimli Sinir Ağı, Uzun Kısa-Süreli Bellek ve Kapılı Tekrarlayan Birim üç öğrenme algoritması ile birlikte BERT, XLNet ve ULMFIT olmak üzere üç transformatör modeli uygulanmıştır. Önerilen modelde Python, Keras API ve Tensorflow birlikte kullanılmıştır. Deneysel sonuçlarda elde edilen performans parametreleri doğruluk, kesinlik, duyarlılık, F1-ölçütü ve AUC olarak belirlenmiş ve LMTweets + CNN modelinin kullanılan tüm modeller arasında daha iyi performans gösterdiği ortaya konmuştur.

🔖

Cite This Article

ÖZBAY, E. (2022). Aggression Detection in Twitter Data Using Transformer‑Based Convolutional Neural Network Model. Konya Mühendislik Bilimleri Dergisi, Cilt: 10 - Sayı: 4, 986–1001. https://ajindex.com/articles/869704

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (7)

  1. Abulaish, M., Kamal, A., Zaki, M., 2020, "A survey of figurative language and its computational detection in online social networks." 14(1): 1-52.
  2. Al-Garadi, M. A., Varathan, K. D., Ravana, S. D. J. C. i. H. B., 2016, "Cybercrime detection in online communications: The experimental case of cyberbullying detection in the Twitter network." 63: 433-443.
  3. Aroyehun, S. T., & Gelbukh, A., 2018. "Aggression detection in social media: Using deep neural networks, data augmentation, and pseudo labeling. In Proceedings of the First Workshop on Trolling, Aggression and Cyberbullying" (TRAC-2018) (pp. 90-97).
  4. Avvaru, A., Vobilisetty, S., & Mamidi, R., 2020, "Detecting sarcasm in conversation context using transformer-based models. In Proceedings of the second workshop on figurative language processing" (pp. 98-103).
  5. Balakrishnan, V., Khan, S., Fernandez, T., Arabnia, H. R. J. P., 2019, "Cyberbullying detection on twitter using Big Five and Dark Triad features." 141: 252-257.
  6. Bansal, A., Sharma, S. M., Kumar, K., Aggarwal, A., Goyal, S., Choudhary, K., 2012, "Classification of flames in computer mediated communications." Basile, V., Bosco, C., Fersini, E., Debora, N., Patti, V., Pardo, F. M. R., 2019, “Semeval-2019 task 5: Multilingual detection of hate speech against immigrants and women in twitter.” 13th International Workshop on Semantic Evaluation, Association for Computational Linguistics. Chatzakou, D., Kourtellis, N., Blackburn, J., De Cristofaro, E., Stringhini, G., & Vakali, A., 2017, “Mean birds: Detecting aggression and bullying on twitter.” Proceedings of the 2017 ACM on web science conference. Chavan, V. S., & Shylaja, S., 2015, “Machine learning approach for detection of cyber-aggressive comments by peers on social media network.” 2015 International Conference on Advances in Computing, Communications and Informatics (ICACCI), IEEE. Chia, Z. L., Ptaszynski, M., & Masui, F., 2019, “Exploring machine learning techniques for irony detection.” Proceedings of the Annual Conference of JSAI 33rd Annual Conference, 2019, The Japanese Society for Artificial Intelligence. Davidson, T., Warmsley, D., Macy, M., & Weber, I., 2017, “Automated hate speech detection and the problem of offensive language.” Proceedings of the International AAAI Conference on Web and Social Media. Dinakar, K., Jones, B., Havasi, C., Lieberman, H., & Picard, R. J. A. T., 2012, "Common sense reasoning for detection, prevention, and mitigation of cyberbullying." 2(3): 1-30.
  7. Djuric, N., Zhou, J.,