Abstract
This paper concentrates on the extraction of pertinent information from unstructured data, specifically analyzing textual content disseminated by users on X/Twitter. The objective is to construct an exhaustive knowledge graph by discerning implicit personal data from tweets. The gleaned information serves to instantiate a digital counterpart and establish a tailored alert mechanism aimed at shielding users from threats such as social engineering or doxing. The study assesses the efficacy of fine-tuning cutting-edge open source large language models for extracting pertinent triples from tweets. Additionally, it delves into the concept of digital counterparts within the realm of cyber threats and presents relevant works in information extraction. The methodology encompasses data acquisition, relational triple extraction, large language model fine-tuning, and subsequent result evaluation. Leveraging a X/Twitter dataset, the study scrutinizes the challenges inherent in user-generated data. The outcomes underscore the precision of the extracted triples and the discernible personal traits gleaned from tweets.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
https://labelstud.io/, accessed: 2024-02-16.
- 2.
The selection of the Miqu-1-70B model for our study is based on its availability as an open-sourced resource, despite its origins in a leak. It’s important to note that, due to these circumstances, there are no official publications or documented references that can be cited. Our choice to utilize this model is rooted in its potential for advancing research objectives, acknowledging the unconventional nature of its dissemination.
- 3.
https://huggingface.co/miqudev/Miqu-1-70B, accessed: 2024-02-16.
- 4.
https://github.com/eth-sri/lmql, accessed: 2024-02-16.
- 5.
https://github.com/guidance-ai/guidance, accessed: 2024-02-16.
- 6.
https://python.langchain.com/docs/guides/evaluation, accessed: 2024-02-16.
References
Ahmed, W., Bath, P.A., Demartini, G.: Using Twitter as a data source: an overview of ethical, legal, and methodological challenges. Ethics Online Res. 2, 79–107 (2017)
Barricelli, B.R., Casiraghi, E., Fogli, D.: A survey on digital twin: definitions, characteristics, applications, and design implications. IEEE Access 7, 167653–167671 (2019). https://doi.org/10.1109/ACCESS.2019.2953499
Bäumer, F.S., Grote, N., Kersting, J., Geierhos, M.: Privacy matters: detecting nocuous patient data exposure in online physician reviews. In: Damaševičius, R., Mikašytė, V. (eds.) ICIST 2017. CCIS, vol. 756, pp. 77–89. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-67642-5_7
Bäumer, F.S., Kersting, J., Orlikowski, M., Geierhos, M.: Towards a multi-stage approach to detect privacy breaches in physician reviews. In: SEMANTICS Posters &Demos (2018)
Bäumer, F.S., Denisov, S., Su Lee, Y., Geierhos, M.: Towards authority-dependent risk identification and analysis in online networks. In: Halimi, A., Ayday, E. (eds.) Proceedings of the IST-190 Research Symposium (RSY) on AI, ML and BD for Hybrid Military Operations (AI4HMO) (2021)
Chan, Y.S., Roth, D.: Exploiting syntactico-semantic structures for relation extraction. In: Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, pp. 551–560 (2011)
Dettmers, T., Pagnoni, A., Holtzman, A., Zettlemoyer, L.: QLoRA: efficient finetuning of quantized LLMs. In: Advances in Neural Information Processing Systems, vol. 36 (2024)
Engels, G.: Der digitale Fußabdruck, Schatten oder Zwilling von Maschinen und Menschen. Gruppe. Interaktion. Organisation. Zeitschrift für Angewandte Organisationspsychologie (GIO) 51(3), 363–370 (2020). https://doi.org/10.1007/s11612-020-00527-9
Feher, K.: Digital identity and the online self: footprint strategies-an exploratory and comparative research study. J. Inf. Sci. 47(2), 192–205 (2021)
Flinn, M.B., Teodorski, C.J., Paullet, K.L.: Raising awareness: an examination of embedded GPS data in images posted to the social networking site twitter. Issues Inf. Syst. 11(1), 432–438 (2010)
Gopi, A.P., Jyothi, R.N.S., Narayana, V.L., Sandeep, K.S.: Classification of tweets data based on polarity using improved RBF kernel of SVM. Int. J. Inf. Technol. 15(2), 965–980 (2020). https://doi.org/10.1007/s41870-019-00409-4
Houlsby, N., et al.: Parameter-efficient transfer learning for NLP. In: International Conference on Machine Learning, pp. 2790–2799. PMLR (2019)
Hu, E.J., et al.: LoRA: low-rank adaptation of large language models. In: International Conference on Learning Representations (2022). https://openreview.net/forum?id=nZeVKeeFYf9
Liu, H., et al.: Few-shot parameter-efficient fine-tuning is better and cheaper than in-context learning. Adv. Neural. Inf. Process. Syst. 35, 1950–1965 (2022)
Mazza, M., Cola, G., Tesconi, M.: Ready-to-(ab)use: from fake account trafficking to coordinated inauthentic behavior on Twitter. Online Soc. Netw. Media 31, 100224 (2022). https://doi.org/10.1016/j.osnem.2022.100224
OpenAI: ChatGPT: Optimizing Language Models for Dialogue (2022). https://openai.com/blog/. Accessed 02 Oct 2023
Schultenkämper, S., Bäumer, F.S.: Privacy risks in German patient forums: a NER-based approach to enrich digital twins. In: Lopata, A., Gudonienė, D., Butkienė, R. (eds.) ICIST 2023. CCIS, vol. 1979, pp. 113–123. Springer, Cham (2024). https://doi.org/10.1007/978-3-031-48981-5_9
Schultenkämper, S., Bäumer, F., Geierhos, M., Lee, Y.S.: From unstructured data to digital twins: from tweets to structured knowledge. In: Proceedings of the Thirteenth International Conference on Social Media Technologies, Communication, and Informatics, SOTICS 2023, pp. 6–11. IARIA (2023)
Shang, Y.M., Huang, H., Sun, X., Wei, W., Mao, X.L.: Relational triple extraction: one step is enough. In: Raedt, L.D. (ed.) Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, IJCAI 2022, pp. 4360–4366. International Joint Conferences on Artificial Intelligence Organization (2022). https://doi.org/10.24963/ijcai.2022/605
Touvron, H., et al.: Llama 2: Open Foundation and Fine-Tuned Chat Models (2023)
Wan, Z., et al.: GPT-RE: In-context Learning for Relation Extraction using Large Language Models (2023)
Xu, X., Zhu, Y., Wang, X., Zhang, N.: How to Unleash the Power of Large Language Models for Few-shot Relation Extraction? (2023)
Zelenko, D., Aone, C., Richardella, A.: Kernel methods for relation extraction. J. Mach. Learn. Res. 3, 1083–1106 (2003)
Zeng, D., Zhang, H., Liu, Q.: Copymtl: copy mechanism for joint extraction of entities and relations with multi-task learning. In: The Thirty-Fourth AAAI Conference on Artificial Intelligence, pp. 9507–9514 (2020)
Zeng, X., Zeng, D., He, S., Liu, K., Zhao, J.: Extracting relational facts by an end-to-end neural model with copy mechanism. In: Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, pp. 506–514 (2018)
Acknowledgments
This research is funded by dtec.bw – Digitalization and Technology Research Center of the Bundeswehr. dtec.bw is funded by the European Union – NextGenerationEU.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Schultenkämper, S., Bäumer, F.S. (2024). Structured Knowledge Extraction for Digital Twins: Leveraging LLMs to Analyze Tweets. In: Phillipson, F., Eichler, G., Erfurth, C., Fahrnberger, G. (eds) Innovations for Community Services. I4CS 2024. Communications in Computer and Information Science, vol 2109. Springer, Cham. https://doi.org/10.1007/978-3-031-60433-1_10
Download citation
DOI: https://doi.org/10.1007/978-3-031-60433-1_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-60432-4
Online ISBN: 978-3-031-60433-1
eBook Packages: Computer ScienceComputer Science (R0)