Demystifying Legalese: An Automated Approach for Summarizing and Analyzing Overlaps in Privacy Policies and Terms of Service
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916214963437568 |
|---|---|
| author | Soneji, Shikha Hoesing, Mitchell Koujalgi, Sujay Dodge, Jonathan |
| author_facet | Soneji, Shikha Hoesing, Mitchell Koujalgi, Sujay Dodge, Jonathan |
| contents | The complexities of legalese in terms and policy documents can bind individuals to contracts they do not fully comprehend, potentially leading to uninformed data sharing. Our work seeks to alleviate this issue by developing language models that provide automated, accessible summaries and scores for such documents, aiming to enhance user understanding and facilitate informed decisions. We compared transformer-based and conventional models during training on our dataset, and RoBERTa performed better overall with a remarkable 0.74 F1-score. Leveraging our best-performing model, RoBERTa, we highlighted redundancies and potential guideline violations by identifying overlaps in GDPR-required documents, underscoring the necessity for stricter GDPR compliance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_13087 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Demystifying Legalese: An Automated Approach for Summarizing and Analyzing Overlaps in Privacy Policies and Terms of Service Soneji, Shikha Hoesing, Mitchell Koujalgi, Sujay Dodge, Jonathan Computation and Language Machine Learning The complexities of legalese in terms and policy documents can bind individuals to contracts they do not fully comprehend, potentially leading to uninformed data sharing. Our work seeks to alleviate this issue by developing language models that provide automated, accessible summaries and scores for such documents, aiming to enhance user understanding and facilitate informed decisions. We compared transformer-based and conventional models during training on our dataset, and RoBERTa performed better overall with a remarkable 0.74 F1-score. Leveraging our best-performing model, RoBERTa, we highlighted redundancies and potential guideline violations by identifying overlaps in GDPR-required documents, underscoring the necessity for stricter GDPR compliance. |
| title | Demystifying Legalese: An Automated Approach for Summarizing and Analyzing Overlaps in Privacy Policies and Terms of Service |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2404.13087 |