TermsBuster: An AI-Powered Smart Privacy Policy and Legal Document Analysis Assistant

TermsBuster is an AI-powered smart privacy policy and legal document analysis assistant developed to make complex legal and privacy-related documents easier to understand at a first level. The system integrates document upload, text extraction, preprocessing, document classification, summarization, TF-IDF-based important-term analysis, and rule-based risk indication within a single Streamlit application.The proposed system reduces the effort required to obtain an initial overview of a lengthy document. Instead of manually searching through a document for its category, important terms, and potentially significant statements, the user can submit the document and receive structured information through the application. The use of TF-IDF provides a method for identifying informative terms, while rule-based analysis provides an understandable approach for highlighting potentially important content.The classification module further improves the organization of the analysis by identifying the general type of document.

[1] M. Bokaie Hosseini, P. K. C, I. Reyes, and S. Egelman, “Identifying and Classifying Third-party Entities in Natural Language Privacy Policies,” Proceedings of the Second Workshop on Privacy in NLP, pp. 18–27, 2020. [2] I. Chalkidis, M. Fergadiotis, P. Malakasiotis, N. Aletras, and I. Androutsopoulos, “LEGAL-BERT: The Muppets Straight Out of Law School,” Findings of the Association for Computational Linguistics: EMNLP 2020, pp. 2898–2904, 2020. [3] D. Mamakas, P. Tsotsi, I. Androutsopoulos, and I. Chalkidis, “Processing Long Legal Documents with Pre-trained Transformers: Modding LegalBERT and Longformer,” Proceedings of the Natural Legal Language Processing Workshop 2022, 2022. [4] N. Limsopatham, “Effectively Leveraging BERT for Legal Document Classification,” Proceedings of the Natural Legal Language Processing Workshop 2021, pp. 210–216, 2021. [5] M. Masala, R. C. A. Iacob, A. S. Uban, M. Cidota, H. Velicu, T. Rebedea, and M. Popescu, “jurBERT: A Romanian BERT Model for Legal Judgement Prediction,” Proceedings of the Natural Legal Language Processing Workshop 2021, pp. 86–94, 2021. [6] Y. Shvartzshanider, A. Balashankar, T. Wies, and L. Subramanian, “Beyond the Text: Analysis of Privacy Statements through Syntactic and Semantic Role Labeling,” Proceedings of the Natural Legal Language Processing Workshop 2023, pp. 85–98, 2023. [7] A. Shukla, P. Bhattacharya, S. Poddar, R. Mukherjee, K. Ghosh, P. Goyal, and S. Ghosh, “Legal Case Document Summarization: Extractive and Abstractive Methods and their Evaluation,” Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing, pp. 1048–1064, 2022. [8] B. Sun and X. Zhou, “ignore at SemEval-2024 Task 5: A Legal Classification Model with Summary Generation and Contrastive Learning,” Proceedings of the 18th International Workshop on Semantic Evaluation (SemEval-2024), pp. 530–535, 2024. [9] E. Lukose, S. De, and J. Johnson, “Privacy Pitfalls of Online Service Terms and Conditions: A Hybrid Approach for Classification and Summarization,” Proceedings of the Natural Legal Language Processing Workshop 2022, pp. 65–75, 2022. [10] This list has a problem: strict alphabetical order would require E. Lukose before I. Chalkidis, so let me give you the genuinely correct final sequence below.