تشخیص قلدری سایبری در شبکه های اجتماعی با یادگیری عمیق مبتنی بر شبکه عصبی CNN و LSTM
محورهای موضوعی : فناوری های نوین در سیستم های توزیع شده و محاسبات الگوریتمی
محسن اقبالی
1
,
کمال میرزائی
2
*
,
رضا عزیزی
3
1 - گروه مهندسی کامپیوتر، واحد میبد، دانشگاه آزاد اسلامی، میبد، ایران
2 - گروه مهندسی کامپیوتر، واحد میبد، دانشگاه آزاد اسلامی، میبد، ایران
3 - گروه مهندسی کامپیوتر، واحد میبد، دانشگاه آزاد اسلامی، میبد، ایران
کلید واژه: شبکه اجتماعی, قلدری سایبری, یادگیری عمیق, شبکه عصبی کانولوشن, شبکه عصبی LSTM,
چکیده مقاله :
یکی از رویکردهای امیدوارکننده در تشخیص زورگویی سایبری استفاده از الگوریتم‌های یادگیری ماشین و یادگیری عمیق است. با این حال، تشخیص آزار سایبری در شبکه های اجتماعی پیچیده است و یک الگوریتم یادگیری ماشین و یادگیری عمیق به تنهایی توانایی زیادی برای تشخیص دقیق زورگویی سایبری ندارند. در این مقاله برای تشخیص زورگویی سایبری در ابتدا با سه روش استخراج ویژگیGloVe ، Word2Vec و TF-IDF ویژگی های اولیه متن استخراج می شود. در مرحله دوم انتخاب ویژگی با استفاده از الگوریتم JSO انجام می شود و در نهایت ویژگی های مهم به عنوان ورودی روش 1DCNN و LSTM در نظر گرفته می شود. آزمایشات در مجموعه داده توئیتر و فیس بوک برای تشخیص زورگویی سایبری انجام می شود. آزمایشات نشان می دهد دقت، حساسیت و صحت روش پیشنهادی در تشخیص زورگویی سایبری در مجموعه داده توئیتر به ترتیب برابر 23/98 درصد، 86/97 درصد و 73/97 درصد است. نتایج نشان می دهد روش پیشنهادی نسبت به روشهای CNN، LSTM و BERT در تشخیص زورگویی سایبری دارای دقت بیشتری است.
One of the promising approaches in cyberbullying detection is machine learning and deep learning algorithms. Nevertheless, detecting cyberbullying in social networks(SN) is complicated. Machine and deep learning methods for detecting cyberbullying in social networks can have a lot of errors. In this manuscript, to detect cyberbullying, the primary features of the text are extracted using three feature extraction methods including GloVe, Word2Vec, and TF-IDF methods. In the second phase, feature selection is done using the JSO algorithm, and finally, the essential features are considered as input to the 1DCNN and LSTM methods. Experiments are conducted on Twitter and Facebook datasets to detect cyberbullying. The results show that the accuracy, sensitivity(recall), and precision of the proposed method(JSO-CNN-LSTM) in detecting cyberbullying on Twitter are 98.23%, 97.86%, and 97.73%. The results demonstrate that the proposed method is more accurate than CNN, LSTM, and BERT methods in detecting cyberbullying.
[1] H. Parlak Sert, & H. Başkale, “Students' increased time spent on social media, and their level of coronavirus anxiety during the pandemic, predict increased social media addiction,” Health Information & Libraries Journal, vol. 40, no. 3, pp. 262-274, 7 Jul 2023, doi: 10.1111/hir.12448.
[2] J. W. Patchin, & S. Hinduja, “Cyberbullying among Asian American youth before and during the COVID‐19 pandemic,” Journal of school health, vol. 93, no. 1, pp. 82-87, 2023, doi: 10.1111/josh.13249.
[3] D. M. H. Kee, A. Anwar, & I. Vranjes, “Cyberbullying victimization and suicide ideation: The mediating role of psychological distress among Malaysian youth,” Computers in Human Behavior, vol. 150, 108000, January 2024, doi: 10.1016/j.chb.2023.108000.
[4] P. J. Macaulay, O. L. Steer, & L. R. Betts, “Bystander intervention to cyberbullying on social media,” In Handbook of Social Media Use Online Relationships, Security, Privacy and Society, vol. 2, pp. 73-99, Academic Press, 2024, doi: 10.1016/B978-0-443-28804-3.00001-6.
[5] E. Mahajan, H. Mahajan, & S. Kumar, “EnsMulHateCyb: Multilingual hate speech and cyberbully detection in online social media,” Expert Systems with Applications, vol. 236, 121228, Feb 2024, doi: 10.1016/j.eswa.2023.121228
[6] S. M. Fati, A. Muneer, A. Alwadain, & A. O. Balogun, “Cyberbullying Detection on Twitter Using Deep Learning-Based Attention Mechanisms and Continuous Bag of Words Feature Extraction,” Mathematics, vol. 11, no. 16, 3567, 15 August 2023 ,doi: 10.3390/math11163567.
[7] C. Iwendi, G. Srivastava, S. Khan, & P. K. R. Maddikunta, “Cyberbullying detection solutions based on deep learning architectures,” Multimedia Systems, vol. 29, no. 3, pp. 1839-1852, June 2023, doi: 10.1007/s00530-020-00701-5.
[8] M. Dadvar, & K. Eckert, “Cyberbullying detection in social networks using deep learning based models. In Big Data Analytics and Knowledge Discovery: 22nd International Conference, DaWaK 2020,” Bratislava, Slovakia, September 14–17, 2020, Proceedings 22, Springer International Publishing, pp. 245-255, Sep 2020, doi: 10.1007/978-3-030-59065-9_20.
[9] A. Bozyiğit, S. Utku, & E. Nasibov, “Cyberbullying detection: Utilizing social media features,” Expert Systems with Applications, vol. 179, 115001, 1 October 2021, doi: 10.1016/j.eswa.2021.115001.
[10] T. Mahmud, M. Ptaszynski, J. Eronen, & F. Masui, “Cyberbullying detection for low-resource languages and dialects: Review of the state of the art,” Information Processing & Management, vol. 60, no. 5, 103454, 27 June 2023, doi: 10.1016/j.ipm.2023.103454.
[11] C. Iwendi, G. Srivastava, S. Khan, & P. K. R. Maddikunta, “Cyberbullying detection solutions based on deep learning architectures,”. Multimedia Systems, vol. 29, no. 3, pp. 1839-1852, June 2023, doi: 10.1007/s00530-020-00701-5.
[12] A. Akhter, U. K. Acharjee, M. A. Talukder, M. M. Islam, & M. A. Uddin, “A robust hybrid machine learning model for Bengali cyber bullying detection in social media,” Natural Language Processing Journal, vol. 4, 100027, September 2023, doi: 10.1016/j.nlp.2023.100027.
[13] H. Saini, H. Mehra, R. Rani, G. Jaiswal, A. Sharma, & A. Dev, “Enhancing cyberbullying detection: a comparative study of ensemble CNN–SVM and BERT models,” Social Network Analysis and Mining, vol. 14, no. 1, 2 December 2023, doi: 10.1007/s13278-023-01158-w.
[14] B. A. H. Murshed, J. Abawajy, S. Mallappa, M. A. N. Saif, & H. D. E. Al-Ariki, “DEA-RNN: A hybrid deep learning approach for cyberbullying detection in Twitter social media platform,” IEEE Access, vol. 10, pp. 25857-25871, 23 February 2022, doi: 10.1109/ACCESS.2022.3153675.
[15] A. Kumar, & N. Sachdeva, “A Bi-GRU with attention and CapsNet hybrid model for cyberbullying detection on social media,” World Wide Web, vol. 25, no. 4, pp. 1537-1550, 01 July 2022, doi: 10.1007/s11280-021-00920-4.
[16] A. Dass, & D. K. Daniel, “Cyberbullying Detection on Social Networks using LSTM Model,” In 2022 International Conference on Innovations in Science and Technology for Sustainable Development (ICISTSD), pp. 293-296, IEEE, August 2020, doi: 10.22214/ijraset.2024.60420.
[17] A. Alam, P. Verma, M. Tariq, A. Sarwar, B. Alamri, N. Zahra, & S. Urooj, “Jellyfish search optimization algorithm for mpp tracking of pv system,” Sustainability, vol. 13, no. 21, 11736, 24 October 2021, doi: 10.3390/su132111736.
[18] A. B. Barragán Martín, M. D. M. Molero Jurado, M. D. C. Pérez-Fuentes, M. D. M. Simon Marquez, Á. Martos Martínez, M. Sisto, & J. J. Gazquez Linares, “Study of cyberbullying among adolescents in recent years: A bibliometric analysis,” International journal of environmental research and public health, vol. 18, no. 6, 3016, 15 March 2021, doi: 10.3390/ijerph18063016.
[19] Á. Denche-Zamorano, S. Barrios-Fernandez, C. Galán-Arroyo, S. Sánchez-González, F. Montalva-Valenzuela, A. Castillo-Paredes, ... & P. R. Olivares, “Science mapping: a bibliometric analysis on cyberbullying and the psychological dimensions of the self,” International journal of environmental research and public health, vol. 20, no. 1, 209, 23 December 2022, doi: 10.3390/ijerph20010209.
[20] M. T. Hasan, M. A. E. Hossain, M. S. H. Mukta, A. Akter, M. Ahmed, & S. Islam, “A Review on Deep-Learning-Based Cyberbullying Detection,” Future Internet, vol. 15, no. 5, 179, 11 May 2023, doi: 10.3390/fi15050179.
[21] C. Iwendi, G. Srivastava, S. Khan, & P. K. R. Maddikunta, “Cyberbullying detection solutions based on deep learning architectures,” Multimedia Systems, vol. 29, no. 3, pp. 1839-1852, June 2023, doi: 10.1007/s00530-020-00701-5.
[22] S. Paul, S. Saha, & J. P. Singh, “COVID-19 and cyberbullying: deep ensemble model to identify cyberbullying from code-switched languages during the pandemic,” Multimedia tools and applications, vol. 82, no. 6, pp. 8773-8789, March 2023, doi: 10.1007/s11042-021-11601-9.
[23] B. A. H. Murshed, Suresha, J. Abawajy, M. A. N. Saif, H. M. Abdulwahab, & F. A. Ghanem, “FAEO-ECNN: cyberbullying detection in social media platforms using topic modelling and deep learning,” Multimedia Tools and Applications, vol. 82, no. 30, pp. 46611–46650, December 2023, doi: 10.1007/s11042-023-15372-3.
[24] V. L. Paruchuri, & P. Rajesh, “CyberNet: a hybrid deep CNN with N-gram feature selection for cyberbullying detection in online social networks,” Evolutionary Intelligence, vol. 16, no. 6, pp. 1935-1949, December 2023, doi: 10.1007/s12065-022-00774-3.
[25] S. Giri, & S. Banerjee, “Performance analysis of annotation detection techniques for cyber-bullying messages using word-embedded deep neural networks,” Social Network Analysis and Mining, vol. 13, no. 23, 14 January 2023, doi: 10.1007/s13278-022-01023-2.
[26] M. Al-Hashedi, L. K. Soon, H. N. Goh, A. H. L. Lim, & E. G. Siew, “Cyberbullying Detection Based on Emotion,” IEEE Access, vol. 11, no. 12, pp. 53907-53918, 29 May 2023, doi: 10.1109/ACCESS.2023.3280556.
[27] N. A. Samee, U. Khan, S. Khan, M. M. Jamjoom, M. Sharif, & D. H. Kim, “Safeguarding Online Spaces: A Powerful Fusion of Federated Learning, Word Embeddings, and Emotional Features for Cyberbullying Detection,” IEEE Access, vol. 11, 2 November 2023, doi: 10.1109/ACCESS.2023.3329347.
[28] A. Muneer, A. Alwadain, M. G. Ragab, & A. Alqushaibi, “Cyberbullying Detection on Social Media Using Stacking Ensemble Learning and Enhanced BERT,” Information, vol. 14, no. 8, 467, 18 August 2023, doi: g/10.3390/info14080467.
[29] A. F. Alqahtani, & M. Ilyas, “An Ensemble-Based Multi-Classification Machine Learning Classifiers Approach to Detect Multiple Classes of Cyberbullying,” Machine Learning and Knowledge Extraction, vol. 6, no. 1, pp.156-170, 12 January 2024, doi: 10.3390/make6010009.
[30] L. Xiaoyan, R. C. Raga, & S. Xuemei, “GloVe-CNN-BiLSTM model for sentiment analysis on text reviews,” Journal of Sensors, vol. 2022, no. 1, 22 October 2022, doi: 10.1155/2022/7212366.
[31] P. Sun, J. Wang, & Z. Dong, “CNN–LSTM Neural Network for Identification of Pre-Cooked Pasta Products in Different Physical States Using Infrared Spectroscopy,” Sensors, vol. 23, no. 10, 4815, 17 May 2023, doi: 10.3390/s23104815.
[32] A. Muneer, & S. M. Fati, “A comparative analysis of machine learning techniques for cyberbullying detection on twitter,” Future Internet, vol. 12, no. 11, 187, 29 October 2020, doi: 10.3390/fi12110187.
[33] S.A.R. Zaidi, Suspicious Communication on Social Platforms. [Online]. Available: https://www.kaggle.com/datasets/syedabbasraza/suspicious-communication-on-social-platforms [Accessed on 20 November 2022].