Summary of Extractive Text Summarisation Of Privacy Policy Documents Using Machine Learning Approaches, by Chanwoo Choi
Extractive text summarisation of Privacy Policy documents using machine learning approaches
by Chanwoo Choi
First submitted to arxiv on: 9 Apr 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper presents two Privacy Policy (PP) summarization models based on K-means clustering and Pre-determined Centroid (PDC) clustering. The first model uses K-means, which outperforms the other nine algorithms in an extensive evaluation. The PDC-based model segregates individual sentences by Euclidean distance from pre-defined cluster centers, defined according to General Data Protection Regulation (GDPR)’s 14 essential topics. The PDC model excels in SSD and ROUGE evaluations, with margins of 27% and 24%, respectively. This contrasts the K-means model’s better performance in general clustering before task-specific evaluation. The results indicate the effectiveness of task-specific fine-tuning measures on unsupervised machine-learning models. The paper demonstrates a method for efficiently extracting essential sentences for PP documents, potentially applicable to GDPR-compliance testing. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper develops two Privacy Policy (PP) summarization models using different clustering algorithms. It compares K-means with Pre-determined Centroid (PDC) clustering and finds that the PDC model performs better in certain evaluations. The PDC model groups sentences by their distance from pre-defined centers based on GDPR topics. This shows how to summarize PP documents efficiently, which could help ensure compliance with data privacy regulations. |
Keywords
* Artificial intelligence * Clustering * Euclidean distance * Fine tuning * K means * Machine learning * Rouge * Summarization * Unsupervised
