Experimenting with Automatic PII Detection on the Hub using Presidio

At Hugging Face, we've noticed a concerning trend in machine learning (ML) datasets hosted on our Hub: Undocumented private information about individuals. This poses some unique challenges for ML practitioners. In this blog post, we'll explore different types of datasets containing a type of private information known as Personally Identifying Information (PII), the issues they present, and a new feature we're experimenting with on the Dataset Hub to help address these challenges.

Types of Datasets with PII

We noticed two types of datasets that contain PII:

1. Annotated PII datasets: Datasets like PII-Masking-300k by Ai4Privacy are specifically designed to train PII Detection Models, which are used to detect and mask PII. For example, these models can help with online content moderation or provide anonymized databases.
2. Pre-training datasets: These are large-scale datasets, often terabytes in size, that are typically obtained through web crawls. While these datasets are generally filtered to remove certain types of PII, small amounts of sensitive information can still slip through the cracks due to the sheer volume of data and the imperfections of PII Detection Models.

The Challenges of PII in ML Datasets

The presence of PII in ML datasets can create several challenges for practitioners. First and foremost, it raises privacy concerns and can be used to infer sensitive information about individuals. Additionally, PII can impact the performance of ML models if it is not properly handled. For example, if a model is trained on a dataset containing PII, it may learn to associate certain PII with specific outcomes, leading to biased predictions or to generating PII from the training set.

A New Experiment on the Dataset Hub: Presidio Reports

To help address these challenges, we're experimenting with a new feature on the Dataset Hub that uses Presidio, an open-source state-of-the-art PII detection tool. Presidio relies on detection patterns and machine learning models to identify PII.