Sign In

Communications of the ACM

ACM TechNews

Researchers Propose Methods for Automatic Detection of Doxing

View as: Print Mobile App Share:
A teen using social media.

A new approach to automatically detecting doxing could lead to more immediate flagging and removal of sensitive personal information that has been shared without the owner’s authorization.

Credit: SolStock/iStock

Pennsylvania State University researchers have developed an automated technique that uses machine learning to identify doxing on social media.

A form of cyberbullying, doxing involves publicly sharing private or personally identifiable information without a person's knowledge or consent.

The researchers established a dataset of close to 180,000 Twitter posts likely featuring doxed information and used machine learning to categorize the data based on a person's identity or their location and then as self-disclosures or malicious disclosures.

The researchers used nine approaches based on existing natural language processing methods and models to determine which was most accurate in automatically detecting doxing involving Social Security numbers and IP addresses in the dataset.

Their proposed approach achieved a more than 96% accuracy rate in automatically detecting doxing on Twitter.

From Pennsylvania State University News
View Full Article


Abstracts Copyright © 2022 SmithBucklin, Washington, DC, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account