Guide

1. Registration

To start using PII Checker you need to register first.

To register fill in the registration form and confirm your email address. Once your email is confirmed we will review your information and approve your account.

2. Maximum package size and usage limits

The maximum size of a package that can be uploaded is 500 MB and the maximum number of data columns for evaluation is 2000.

If your package is too large please delete any files that do not need to be checked.

You are allowed to process maximum of 5 packages a day.

These limits are implemented to enable the maximum number of researchers to use this service.

If you need to process more, or larger, packages please contact us and we can adjust the limits.

3. Uploading a package

From your dashboard, choose a file and select Upload. The uploaded file can be a full package as an archive (.zip) or an individual data file. The uploaded file's name is shown immediately in your package list, with status Queued.

4. What happens after upload

Processing has two phases, both visible in your package status without any action required from you:

  1. Extraction and screening. The package is unpacked, duplicate files are identified, and every column of every data file is screened to decide which columns need to be checked. A package that has over 2000 candidate columns is rejected outright at this stage.
  2. Classification. Each candidate column's values are tabulated and classified. Your dashboard shows live progress as Running — X of Y columns processed… once the total is known.

5. Supported file formats

PII Checker is designed to evaluate tabular data. It does not search for PII in documents such as .pdf and .doc or in multimedia files. Currently the following formats are supported:

Supported data formats are: Excel (.xls, .xlsx), Plain text tabular data (.csv, .tsv, .txt), Stata (.dta), R (.rds, .rdata), SPSS (.sav), SAS (.sas7bdat, .xpt), Matlab (.mat), and OpenDocument Spreadsheet (.ods).

6. Report

Once the

Each row in the downloaded report corresponds to one column found in a dataset in the package:

Column Meaning
file Path of the file the column was found in, relative to the package
sheet Sheet name, for multi-sheet formats (Excel, etc.)
col_name The column's variable name.
label The column's descriptive label for formats that support labels
n_rows Number of non-missing values in the column
evaluation The classification: direct_pii, possible_indirect, internal_id, consent or not_pii
reasoning The model's stated reason for that classification
duplicate_of If this file was a duplicate of another file in the package, the original's path

7. Classification legend

Each candidate data column can be classified as one of the following:

7. Classification legend

Each candidate column is assigned one of the following classifications:

Classification Meaning
direct_pii The column directly identifies an individual — for example, a name, email address, phone number, precise location, or IP address.
possible_indirect The column could help identify an individual when combined with other data — for example, a village name, a detailed demographic characteristic, or a rare occupation.
internal_id The column is a unique internal identifier, such as a respondent ID, household ID, or case number.
consent The column records whether a respondent provided consent, or details relating to consent.
not_pii The column does not seem to identify individuals or fit any of the other categories.

8. Naming a package

By default a package is identified by the file name you uploaded.

You can give your package a more memorable name by typing it into the textbox and confirm it by clicking the checkmark button. The name you choose is also used to name the downloaded report file.

To edit the name of your package click on the pencil icon.

9. Cancelling a running job

Select Cancel on a package that is actively running to stop it. A cancelled job produces no report — there is no partial download.

10. Deleting a package or report

Select Delete on any package to remove its job record and any files still associated with it.