Logo PTI Logo FedCSIS

Proceedings of the 19th Conference on Computer Science and Intelligence Systems (FedCSIS)

Annals of Computer Science and Information Systems, Volume 39

Impact of Spelling and Editing Correctness on Detection of LLM-Generated Emails

, , , ,

DOI: http://dx.doi.org/10.15439/2024F8906

Citation: Proceedings of the 19th Conference on Computer Science and Intelligence Systems (FedCSIS), M. Bolanowski, M. Ganzha, L. Maciaszek, M. Paprzycki, D. Ślęzak (eds). ACSIS, Vol. 39, pages 603–608 ()

Full text

Abstract. In this paper, we investigated the impact of spelling and editing correctness on the accuracy of detection if an email was written by a human or if it was generated by a language model. As a dataset, we used a combination of publicly available email datasets with our in-house data, with over 10k emails in total. Then, we generated their``copies'' using large language models (LLMs) with specific prompts. As a classifier, we used random forest, which yielded the best results in previous experiments. For English emails, we found a slight decrease in evaluation metrics if error-related features were excluded. However, for the Polish emails, the differences were more significant, indicating a decline in prediction quality by around 2\% relative. The results suggest that the proposed detection method can be equally effective for English even if spelling- and grammar-checking tools are used. As for Polish, to compensate for error-related features, additional measures have to be undertaken.

References

  1. Europol, “Chatgpt: the impact of large language models on law enforcement,” 2023. http://dx.doi.org/10.2813/255453
  2. H. Williams and C. McCulloch, “Truth decay and national security: Intersections, insights, and questions for future research,” Santa Monica, CA, USA, 2023. [Online]. Available: https://www.rand.org/pubs/perspectives/PEA112-2.html
  3. K. T. Gradoń, “Generative artificial intelligence and medical disinformation,” British Medical Journal, no. 384, 2024. http://dx.doi.org/10.1136/bmj.q579
  4. P. Gryka, K. Gradoń, M. Kozłowski, M. Kutyła, and A. Janicki, “Detection of AI-generated emails – a case study,” in Proc. 13th International Workshop on Cyber Crime (IWCC 2024), Vienna, Austria, 2024, (accepted for publication).
  5. A. Knott, D. Pedreschi, R. Chatila, T. Chakraborti, S. Leavy, R. Baeza-Yates, D. Eyers, A. Trotman, P. D. Teal, P. Biecek, S. Russell, and Y. Bengio, “Generative AI models should include detection mechanisms as a condition for public release,” Ethics and Information Technology, vol. 25, no. 4, p. 55, 12 2023. http://dx.doi.org/10.1007/s10676-023-09728-4
  6. K. Krishna, Y. Song, M. Karpinska, J. Wieting, and M. Iyyer, “Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defense,” Advances in Neural Information Processing Systems, vol. 36, 3 2024.
  7. F. Jelinek, R. L. Mercer, L. R. Bahl, and J. K. Baker, “Perplexity—a measure of the difficulty of speech recognition tasks,” The Journal of the Acoustical Society of America, vol. 62, no. S1, pp. S63–S63, 1977. http://dx.doi.org/10.1121/1.2016299
  8. M. Chakraborty, S. T. I. Tonmoy, S. M. M. Zaman, S. Gautam, T. Kumar, K. Sharma, N. Barman, C. Gupta, V. Jain, A. Chadha, A. Sheth, and A. Das, “Counter Turing test (CT2): AI-generated text detection is not as easy as you may think - introducing AI detectability index (ADI),” in Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, H. Bouamor, J. Pino, and K. Bali, Eds. Singapore: Association for Computational Linguistics, Dec. 2023. http://dx.doi.org/10.18653/v1/2023.emnlp-main.136 pp. 2206–2239. [Online]. Available: https://aclanthology.org/2023.emnlp-main.136
  9. I. Cingillioglu, “Detecting AI-generated essays: the ChatGPT challenge,” International Journal of Information and Learning Technology, vol. 40, pp. 259–268, 5 2023. http://dx.doi.org/10.1108/IJILT-03-2023-0043
  10. L. Fröhling and A. Zubiaga, “Feature-based detection of automated language models: tackling GPT-2, GPT-3 and Grover,” PeerJ Computer Science, vol. 7, p. e443, 4 2021. http://dx.doi.org/10.7717/peerj-cs.443
  11. Y. Shi, Q. Sheng, J. Cao, H. Mi, B. Hu, and D. Wang, “Ten words only still help: Improving black-box AI-generated text detection via proxy-guided efficient re-sampling,” arXiv preprint, vol. https://arxiv.org/abs/2402.09199, 2024. [Online]. Available: http://arxiv.org/abs/2402.09199
  12. E. Mitchell, Y. Lee, A. Khazatsky, C. D. Manning, and C. Finn, “DetectGPT: Zero-shot machine-generated text detection using probability curvature,” in Proc. International Conference on Machine Learning. Online: PMLR, 2023, pp. 24 950–24 962.
  13. F. Harrag, M. Dabbah, K. Darwish, and A. Abdelali, “Bert transformer model for detecting Arabic GPT2 auto-generated tweets,” in Proceedings of the Fifth Arabic Natural Language Processing Workshop, I. Zitouni, M. Abdul-Mageed, H. Bouamor, F. Bougares, M. El-Haj, N. Tomeh, and W. Zaghouani, Eds. Barcelona, Spain (Online): Association for Computational Linguistics, Dec. 2020, pp. 207–214. [Online]. Available: https://aclanthology.org/2020.wanlp-1.19
  14. J. D. Rodriguez, T. Hay, D. Gros, Z. Shamsi, and R. Srinivasan, “Cross-domain detection of GPT-2-generated technical text,” in Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, vol. 2022.naacl-main.88. Association for Computational Linguistics, 2022. http://dx.doi.org/10.18653/v1/2022.naacl-main.88 pp. 1213–1233.
  15. S. Mukherjee, “Exploring burstiness: Evaluating language dynamics in LLM-generated texts,” 2023, [Online]. Available: https://ramblersm.medium.com/exploring-burstiness-evaluating-language-dynamics-in-llm-generated-texts-8439204c75c1 (Accessed on Apr 30, 2024).
  16. I. Okulska, D. Stetsenko, A. Kołos, A. Karlińska, K. Głąbińska, and A. Nowakowski, “Stylometrix: An open-source multilingual tool for representing stylometric vectors,” arXiv preprint https://arxiv.org/abs/2309.12810, vol. 2309.12810, 9 2023.
  17. J. Morris, “LanguageTool Python library,” 2024, https://pypi.org/project/language-tool-python/ (Accessed on May 10, 2024). [Online]. Available: https://pypi.org/project/language-tool-python/
  18. F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay, “Scikit-learn: Machine learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011.
  19. _w1998, “Spam email dataset,” 2023, (Accessed on Jan 14, 2024). [Online]. Available: https://www.kaggle.com/datasets/jackksoncsie/spam-email-dataset/data
  20. R. Modi, “Email classification dataset,” 2023, (Accessed on Jan 14, 2024). [Online]. Available: https://github.com/rmodi6/Email-Classification/tree/master
  21. Apache Public Datasets, “The Spam Assassin Email Classification Dataset,” 2023, (Accessed on Jan 14, 2024). [Online]. Available: https://www.kaggle.com/datasets/ganiyuolalekan/spam-assassin-email-classification-dataset/data