What is it about?

This paper explains how people can be identified from online shopping records even when names are removed. Researchers studied a popular website where people buy and sell used items. They found that item titles, prices, and shopping times can act like fingerprints. When combined, these small clues make it easy to single out one person from many. The study also shows how computer programs can read these clues and build fake profiles that look real. This means that simple hiding of names is not enough to keep shoppers safe.

Featured Image

Why is it important?

Many websites share shopping data to help researchers build better recommendation systems. This paper shows that sharing this data is riskier than people thought. The findings matter now because more websites are releasing large datasets without strong protections. The study gives website owners clear steps to improve privacy, such as removing item titles and making time records less exact. Without these changes, people who think they are anonymous could still be identified and tracked.

Perspectives

As a researcher, I believe this work fills a gap that many people ignore. Most privacy studies focus on social media or health records. Shopping data seems harmless, but it carries rich details about daily life, hobbies, and habits. I hope this paper pushes websites to treat shopping behavior as sensitive information. The fake profiles we built show that even small data leaks can be turned into personal stories. This should alarm anyone who cares about online privacy.

Sami El Yaagoubi
Florida Polytechnic University

Read the Original

This page is a summary of: STALKER: Composed Quasi-Identifiers and LLM-Amplified Behavioral Leakage in Anonymized Marketplace Data, June 2026, ACM (Association for Computing Machinery),
DOI: 10.1145/3806007.3810967.
You can read the full text:

Read

Contributors

The following have contributed to this page