Environment and Social Psychology

Generating culturally-contextual chinese cyberbullying datasets: A GAN approach for social psychology research

Xiangqin Dai

1 School of Computer Sciences, Universiti Sains Malaysia, Pulau, Pinang, 11800, Malaysia 2 Key Laboratory of Intelligent Information Processing and Control, Chongqing Three Gorges University, Chongqing, 404000, China

Mohd Najwadi Yusoff

School of Computer Sciences, Universiti Sains Malaysia, Pulau, Pinang, 11800, Malaysia

Bingli Zhu

Key Laboratory of Intelligent Information Processing and Control, Chongqing Three Gorges University, Chongqing, 404000, China

Xiao Zhang

School of Computer Sciences, Universiti Sains Malaysia, Pulau, Pinang, 11800, Malaysia

Wujin Jiang

Key Laboratory of Intelligent Information Processing and Control, Chongqing Three Gorges University, Chongqing, 404000, China

Lei Wang

School of Computer Sciences, Universiti Sains Malaysia, Pulau, Pinang, 11800, Malaysia

DOI: https://doi.org/10.59429/esp.v10i7.3834

Keywords: generative adversarial networks; Chinese cyberbullying dataset; LeakGAN; cyberbullying

---

Abstract

Cyberbullying has become a growing concern with serious psychological and social consequences, including anxiety, depression, and disrupted online communities. Grounded in social psychology theories such as social learning and online disinhibition, cyberbullying is shaped by factors like anonymity and peer influence. However, the lack of Chinese-language cyberbullying datasets limits research and intervention efforts. To address this, we used four GAN models SeqGAN, RankGAN, MaliGAN, and LeakGAN to generate realistic Chinese cyberbullying text. LeakGAN outperformed the others, achieving a BLEU2 score of 0.948, self-BLEU2 of 0.963, NLL of 0.48, and the highest EmbSim values. Beyond technical performance, we emphasized psychological validity, cultural relevance, and ethical considerations in the data generation process. The findings have important implications for automated detection, intervention design, and social psychology research. Framed within ecological systems theory, this work also considers how online environments shape behavior. The synthetic dataset supports applications in schools, workplaces, and cross-cultural studies, though limitations remain in capturing the full complexity of real human behavior. Overall, LeakGAN’s success offers a strong foundation for future research on cyberbullying in digital contexts.

---

References

[1]. 1.Smith, P. K., Mahdavi, J., Carvalho, M., Fisher, S., Russell, S., and Tippett, N. (2008). Cyberbullying: its nature and impact in secondary school pupils. J. Child Psychol. Psychiatry 49, 376–385. doi:10. 1111/j.1469-7610.2007.01846.x

[2]. 2.Kaimba Frank.School intervention in peer cyberbullying: the case of high school students in Kafue District, Zambia [D]. Supervisor: MingHuo. Northeast Normal University,2023

[3]. 3.Ding Haoran. Research on Cyberbullying Problem of Minors and Its Governance [D]. Supervisor: Xu Lei. Nanjing University of Posts and Telecommunications,2023.

[4]. 4.Fu Brazier. An international comparative study of cyberbullying prevention programmes in primary schools[D]. Supervisor: QianSongling;Men Xinwei. Jilin University of Foreign Languages,2023.

[5]. 5.65.3% of surveyed youth said they or people around them have experienced cyber violence[EB/OL].(2023-6-20) https://baijiahao.baidu.com/s?id= 1769178755106899787&wfr=spider&for=pc

[6]. 6.Rao, M. E., and Rao, D. M. (2021). The mental health of high school students during the COVID- 19 pandemic. Front. Educ. 6, 719539. doi:10.3389/feduc.2021.719539

[7]. 7.Englander E. (2021). 3.5 social and mental health during the COVID- 19 ,pandemic. J. Am. Acad. Child Adolesc. Psychiat. 60, S147. doi: 10. 1016/j.jaac.2021. 09.039

[8]. 8.Paat, Y. F., and Markham, C. (2020). Digital crime, trauma, and abuse: Internet safety and cyber risks for adolescents and emerging adults in the 21st century. Soc. Work Ment. Health 19, 18–40. doi:10. 1080/15332985.2020.1845281

[9]. 9.Kim, Y. J., Qian, L., and Aslam, M. S. (2020). Development of a personalized mobile mental health intervention for workplace cyberbullying among health practitioners: protocol for a mixed methods study. JMIR Res. Protoc. 9, e23112. doi: 10.2196/23112

[10]. 10.Nochaiwong, S., Ruengorn, C., Thavorn, K., Hutton, B., Awiphan, R., Phosuya, C., et al. (2021). Global prevalence of mental health issues among the general population during the coronavirus disease-2019 pandemic: a systematic review and meta-analysis. Sci. Rep. 11, 1. doi:10. 1038/s41598-021-89700-8

[11]. 11.Kowalski, R. M., Toth, A., and Morgan, M. (2017). Bullying and cyberbullying in adulthood and the workplace. J. Soc. Psychol. 158,64–81. doi:10. 1080/00224545.2017.1302402

[12]. 12.Kowalski, R. M., Limber, S. P., and McCord, A. (2018). A developmental approachto cyberbullying: prevalence and protective factors. Aggress. Violent Behav. 45, 20–32.doi:10. 1016/j.avb.2018.02.009

[13]. 13.Goodfellow, I. J., et al. (2014). Generative Adversarial Nets. Advances in Neural Information Processing Systems.

[14]. 14.Andreini, P., Ciano, G., Bonechi, S., Graziani, C., Lachi, V., Mecocci, A., ... & Bianchini, M. (2021). A two-stage GAN for high-resolution retinal image generation and segmentation. Electronics, 11(1), 60.

[15]. 15.Huang, G., & Jafari, A. H. (2023). Enhanced balancing GAN: Minority-class image generation. Neural computing and applications,35(7), 5145-5154.

[16]. 16.Tran, N. T., Tran, V. H., Nguyen, N. B., Nguyen, T. K., & Cheung, N. M. (2021). On data augmentation for GAN training. IEEE Transactions on Image Processing, 30, 1882- 1897.

[17]. 17.Liu, Y. (2021). Improved generative adversarial network and its application in image oil painting style transfer. Image and Vision Computing, 105, 104087.

[18]. 18.Chen, Y., Zhang, H., Liu, L., Chen, X., Zhang, Q., Yang, K., ... & Xie, J. (2021). Research on image inpainting algorithm of improved GAN based on two-discriminations networks. Applied Intelligence, 51,3460-3474.

[19]. 19.Zhang, B., Gu, S., Zhang, B., Bao, J., Chen, D., Wen, F., ... & Guo, B. (2022). Styleswin: Transformer-based gan for high-resolution image generation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 11304- 11314).

[20]. 20.Singh, N. K., & Raza, K. (2021). Medical image generation using generative adversarial networks: A review. Health informatics: A computational perspective in healthcare, 77-96.

[21]. 21.Hossam, M., Le, T., Papasimeon, M., Huynh, V., & Phung, D. (2021). Text generation with deep variational GAN. arXiv preprint arXiv:2104.13488.

[22]. 22.Chen, Z., Zhu, T., Xiong, P., Wang, C., & Ren, W. (2021). Privacy preservation for image data: a gan‐based method. International Journal of Intelligent Systems, 36(4), 1668- 1685.

[23]. 23.Yu, L., Zhang, W., Wang, J., & Yu, Y. (2017, February). Seqgan: Sequence generative adversarial nets with policy gradient. In Proceedings of the AAAI conference on artificial intelligence (Vol. 31, No. 1).

[24]. 24.Guo, J., Lu, S., Cai, H., Zhang, W., Yu, Y., & Wang, J. (2018, April). Long text generation via adversarial training with leaked information. In Proceedings of the AAAI conference on artificial intelligence (Vol. 32, No. 1).

[25]. 25.Ding, X., Wang, Y., Xu, Z., Welch, W. J., & Wang, Z. J. (2021, May). Ccgan: Continuous conditional generative adversarial networks for image generation. In International conference on learning representations.

[26]. 26.Diao, S., Shen, X., Shum, K., Song, Y., & Zhang, T. (2021, August). TILGAN: transformer-based implicit latent GAN for diverse and coherent text generation. In Findings of the Association for Computational linguistics: ACL-IJCNLP 2021 (pp. 4844-4858).

[27]. 27.Yang, Y., Dan, X., Qiu, X., & Gao, Z. (2020). FGGAN: Feature-guiding generative adversarial networks for text generation. IEEE Access, 8, 105217- 105225.

[28]. 28.Deng Yang, Gao Kun, Liao Ning, Chen Yiran. Automatic script generation and optimisation technique based on generative adversarial network[J]. Digital Technology and Application,2024,42(02):232-234.

[29]. 29.LI Bing,YANG Peng,SUN Yuankang,HU Zhongjian,YI Meng. Advances and Challenges in Artificial Intelligence Text Generation (in English)[J].Frontiers of Information Technology & Electronic Engineering,2024,25(01):64-84.

[30]. 30.Xiong Lu,Pei Zhili,Jiang Mingyang & Bao Qiming. (2023). A text generation model based on improved generative adversarial network. Journal of Inner Mongolia University for Nationalities (Natural Science Edition) (02), 118- 123. doi:10. 14045/j.cnki.15- 1220.2023.02.005.

[31]. 31.Peng, P. F. & Zhou, L. R.. (2022). Adding reward to GRU adversarial network text generation model. Computers and Modernisation (07), 121- 126. doi:CNKI:SUN:JYXH.0.2022-07-019

[32]. 32.Tingting Zhao, Yajing Song, Guixi Li, Lina Wang, Yarui Chen, Dehua Ren. A review of text generation research based on deep reinforcement learning[J]. Journal of Tianjin University of Science and Technology,2022,37(02):71-80.DOI:10. 13364/j.issn.1672-6510.20210146.

[33]. 33.Q. Xue, X. F. Meng, F. Zhang, X. Y. Zhang, J. M. Zhu, Y. Zhu & D. D. Wang. (2022).HLMGAN: Hierarchical learning for multi-reward text generation adversarial networks. Journal of Yunnan University (Natural Science Edition)(01),64-72.

[34]. 34.Y. Y. Kang, D. L. Peng, Z. Chen & C. C. Liu. (2019).ED-GAN: A legal text generation model based on improved generative adversarial networks. Small Microcomputer Systems (05), 1020- 1025. doi:CNKI:SUN:XXWX.0.2019-05-021.

[35]. 35.Zhou, W., Ge, T., Xu, K., Wei, F., & Zhou, M. (2020). Self-adversarial learning with comparative discrimination for text generation. arXiv preprint arXiv:2001.11691.

[36]. 36.Mikolov T, Martin Karafiát, Burget L, et al.Recurrent neural network based language model[C]//Interspeech, Conference of the International Speech Communication Association, Makuhari, Chiba, Japan, September.DBLP, 2015.DOI:10. 1109/EIDWT.2013.25.

[37]. 37.Yu L, Zhang W, Wang J, et al. Seqgan: Sequence generative adversarial nets with policy gradient[C]//Proceedings of the AAAI conference on artificial intelligence. 2017, 31(1).

[38]. 38.Che T, Li Y, Zhang R, et al. Maximum-likelihood augmented discrete generative adversarial networks[J]. arXiv preprintarXiv:1702.07983, 2017.

[39]. 39.Lin K, Li D, He X, et al. Adversarial ranking for language generation[J]. Advances in neural information processing systems, 2017,30.

[40]. 40.Guo J, Lu S, Cai H, et al. Long text generation via adversarial training with leaked information[C]//Proceedings of the AAAI conference on artificial intelligence. 2018, 32(1).

[41]. 41.Zhu Y, Lu S, Zheng L, et al. Texygen: A benchmarking platform for text generation models[C]//The 41st international ACM SIGIR conference on research & development in information retrieval. 2018:1097- 1100.

[42]. 42.Papineni K, Roukos S, Ward T, et al. Bleu: a method for automatic evaluation of machine translation[C]//Proceedings of the 40th annual meeting of the Association for Computational Linguistics. 2002: 311-318.