There is a critical need for a solution to protect sensitive data in an efficient, easy and automated manner. It is a common practice in various industries to obfuscate documents, however, existing software such as Adobe Acrobat Professional, ObjectiveRedact, and PDF element is either expensive, complex to use, or do not provide sufficient redaction features. The goal of our research is to create an automated redaction tool that can automatically detect and remove sensitive data in Word (.docx) and PDF (.pdf) documents. The tool will utilize pattern recognition algorithms and natural language processing techniques to redact sensitive information including custom user sensitive information. Minimizing the risk of data leakage, saving time, ensuring compliance with regulatory requirements, and allowing users to define their own custom sensitive data patterns to redact are some of the primary objectives of the tool. The tool also offers the users a preview and verification method which allows users to examine and improve redaction suggestions for better accuracy. This feature also promotes keeping the human element in the loop. Robust encryption methods have been put in place to safeguard documents that are being stored.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

SquidAct: Automated Custom Redaction

  • Emil Pulickal,
  • Yashesh Kirtikar,
  • Tanvi Dalvi,
  • Ashwini Rao

摘要

There is a critical need for a solution to protect sensitive data in an efficient, easy and automated manner. It is a common practice in various industries to obfuscate documents, however, existing software such as Adobe Acrobat Professional, ObjectiveRedact, and PDF element is either expensive, complex to use, or do not provide sufficient redaction features. The goal of our research is to create an automated redaction tool that can automatically detect and remove sensitive data in Word (.docx) and PDF (.pdf) documents. The tool will utilize pattern recognition algorithms and natural language processing techniques to redact sensitive information including custom user sensitive information. Minimizing the risk of data leakage, saving time, ensuring compliance with regulatory requirements, and allowing users to define their own custom sensitive data patterns to redact are some of the primary objectives of the tool. The tool also offers the users a preview and verification method which allows users to examine and improve redaction suggestions for better accuracy. This feature also promotes keeping the human element in the loop. Robust encryption methods have been put in place to safeguard documents that are being stored.