Parse a text
This helper is designed for extracting information from structured, templated text inputs. Define fields in the structured text to extract and use in your workflow.
Parse a text is a sophisticated tool designed for extracting information from structured text inputs, like invoices, receipts, notifications, and more. By allowing users to define templates for parsing, this helper can identify and label specific data within the text, making it invaluable for data extraction and processing tasks.
It works best if the text follows an identifiable template or pattern.
Common use-cases
Extracting structured data from emails, such as order numbers, shipping details, or contact information.
Processing natural language inputs to retrieve specific information like dates, names, or locations.
Automating data entry from text sources into databases or spreadsheets.
Step-by-step instructions
Text input: Begin by entering the text from which you wish to extract information.
Highlight and label data: Use the template to identify and categorize specific pieces of data within the text. This process involves selecting portions of the text and assigning them labels based on the predefined template.
Example 1 - Data extraction
Let's say you need to extract customer information from a inquiry form submitted via email. The email bodies follow a template with identifiable parts for name, email, and inquiry.
Map email body to input: Name: John Doe, Email: john.doe@example.com, Inquiry: Looking for pricing information on product #2345.
Create a template that identifies and labels the following pieces of data by highlighting and tagging:
Highlight “John Doe”, and tag “Name”
Highlight “john.doe@example.com”, and tag “Email”
Highlight “Looking for pricing information on product #2345”, and tag “Inquiry”
Data extraction: Applying the Parse a text helper with your template to the text input would result in the following structured output:
Last updated