Anonymize Text
Mask/hide personally identifiable information (PII) in text
Introduction
Welcome to the user guide for the "Anonymize Text" Workflow! This Workflow is designed to help you protect sensitive information by masking or hiding personally identifiable information (PII) in text documents. Whether you are a market researcher, customer support reporter, data privacy officer, content creator, or simply concerned about safeguarding personal data, this Workflow is here to assist you. With its user-friendly interface and powerful anonymization capabilities, you can ensure the privacy and security of sensitive information with ease.
Overview
The "Anonymize Text" Workflow is your go-to solution for anonymizing text documents. It utilizes advanced NLP and large language models to detect and mask personally identifiable information (PII) in your text data. By replacing sensitive data with generic placeholders or pseudonyms, you can protect personal information and comply with data protection regulations. With its intuitive design and efficient anonymization process, this Workflow is a must-have for anyone dealing with sensitive information.
Key Features
PII Detection: The Workflow automatically identifies personally identifiable information (PII) in your text documents. It recognizes sensitive data such as names, addresses, phone numbers, and email addresses. This feature ensures that all PII is properly identified for anonymization.
How to Use the Workflow
Locate the Workflow in the template page and click on Use template. You can use the Workflow as is or clone it.
Provide the Text: Use copy-paste to provide the input. Or when in a bulk-run, select the corresponding column containing the data to be anonymized.
Run the Workflow: Once you have provided the input, click the “Run Workflow” button (on the App page) or use the run options on your data table (bulk/single run) to initiate the analysis process. The Workflow will automatically detect and mask the personally identifiable information (PII) in your documents.
View Results: The Workflow will provide you with the PII-masked version of the input.
Even though our tests have shown a high success rate in anonymization, we highly recommend checking the results to ensure safeguarding personal data.
Workflow Execution
Workflows and templates can be:
Tested on individually provided inputs:
Single run on the App page
Single run on the Build page
Single run on the data table
Set to fetch the data from a dataset and apply the analysis on the whole dataset:
Bulk run on the data table
Deep Dive into the Workflow
Workflow Components
If you clone a template or create a Workflow from scratch, you will have access to the Build tab. Build is where you put together different components to build a Workflow suitable for your needs.
User Inputs
Long Text Input: An input text component suitable for long text pieces (more than one line), such as documents, reports, or feedback text.
Actions
Large Language Model (LLM): A large language model component is all set up to provide you access to GPT (and many other LLMs). In the prompt section, you will provide the required information as well as instructions on what is expected to be done.
A Good Prompt
Be short and precise with your instruction/request from the LLM.
Explicitly note constraints and goals.
Include a few examples when possible.
Example Prompt
Running the Workflow
After setting up the Workflow, you can test it with sample data to ensure it works as expected. Once satisfied with the results, you can run it on larger datasets in bulk.
Saving and Exporting Results
Once the anonymization process is complete, you can view the results directly on the data table. You can also export the anonymized data as needed.
This comprehensive guide should help you effectively use the "Anonymize Text" Workflow to protect sensitive information in your text documents.
Last updated