*Result*: Enhanced personally identifiable information data masker using natural language processing and computer vision.

Title:
Enhanced personally identifiable information data masker using natural language processing and computer vision.
Authors:
Hajari, Manu1 (AUTHOR) manu.hajari@gmail.com, Balbudhe, Aniruddha Anil1 (AUTHOR) aniruddha2846@gmail.com, Annavarapu, Tejesh1 (AUTHOR) tejeshannavarapu1804@gmail.com, Mood, Shashikanth1 (AUTHOR) shashichawan790@gmail.com, Maithili, Kamalakannan2 (AUTHOR), Aljanabi, Saif Obyd3 (AUTHOR), Mishra, Sanjay4 (AUTHOR), Bhalla, Lalit5 (AUTHOR)
Source:
AIP Conference Proceedings. 2025, Vol. 3263 Issue 1, p1-8. 8p.
Database:
Academic Search Index

*Further Information*

*The "Enhanced PII Data Masker" work outlines a solution that utilizes NLP and CV to change the approach in security concerning documents. It employs Python, spaCy, OpenCV, and Django to detect and redact sensitive data with high accuracy irrespective of the document's format. Having provided various examples of generic and specific solutions, assessment of experiment results proves the effectiveness of the system to protect sensitive data and, consequently, augment confidentiality and privacy of most digital documents in a diverse range of domains. Farther to extend the functionality of the system, some features are added such as Usability of GUI interface is improved, Feedback mechanism for fine tuning the redaction algorithm, Batch processing. All these improvements combined give a solid answer to not only nuances recognition of private information but also its uncompromised occlusion, which is engaging two important imperatives of the modern informational security. [ABSTRACT FROM AUTHOR]*