Flow Blocks

Flow Blocks

With the introduction of flows, we have created the following types of blocks to help you customize your flows to meet your teams’ needs. To learn more about any of these blocks, or for assistance in adding them to your flows, contact your Hyperscience representative.

Block types and settings

Input Blocks

Formerly known as “Input Connectors,” Input Blocks allow you to integrate your organization’s data sources into our system. Through these blocks, you can process documents from a variety of sources, such as inboxes, message queues, or folders on the network.

A full list of the Input Blocks we currently support, along with more details about each, can be found in Input Blocks.

Settings

Name Required? Description
Allow API submissions No Indicates whether the flow accepts submissions submitted via API.
Allow manual submissions No Indicates whether the flow accepts manually uploaded submissions.

File Filter Block

If your submissions contain files that should not be processed by Hyperscience (e.g., images in email signatures), you can configure the File Filter Block to ensure that only the files that require processing are included in submissions.

You can use the File Filter Block to filter files from submissions that are created through the following methods:

Settings

Name Required? Description
All Files or Images Only Yes Determines whether the filters in this block are applied to all files in submissions or only to image files (i.e., files whose MIME type is image).
Minimum Image Width (px) No The minimum width in pixels that an image needs to have in order to be allowed by the filter.
Minimum Image Height (px) No The minimum height in pixels that an image needs to have in order to be allowed by the filter.
Minimum File Size (KB) No The minimum size in kilobytes that a file needs to have in order to be allowed by the filter.
File Extension Action Yes Select one of the following options:

Submission Initialization

The Submission Initialization Block contains settings that connect your flow to your:

AWS S3 Settings

S3 Submission Retrieval Store

If you are using an S3 bucket as your submission retrieval store and you are not authenticating through IAM roles, provide your AWS access key ID and secret access key in the S3 Submission Retrieval Store field.

To enter your credentials:

  1. Click Edit value.
  2. Enter your credentials in JSON format:
{
  "aws_access_key_id": "<your_access_key_id>",
  "aws_secret_access_key": "<your_secret_key>"
}

S3 Submission Retrieval Endpoint URL

If your submission retrieval store is not in the public cloud, enter its URL in S3 Submission Retrieval Endpoint URL.

Classification Blocks

We’ve divided our Classification function into two blocks: one for Machine Classification and another for Manual Classification.

Machine Classification

With machine classification, Hyperscience can automatically match your submissions to Structured, Semi-structured, or Additional layouts.

Settings

Name Required? Description
Image Correction No Identifies and corrects the orientation of Semi-structured images by rotating them.
Faster PDF Transcription No If enabled, the system processes pages in PDF files in their native format, allowing for faster transcription.
Captured Image Enhancement No Improves machine readability of Semi-structured documents captured by mobile devices.

Manual Classification

Manual Classification, or Classification Supervision, allows your keyers to manually match submissions to their layouts.

Settings

Name Required? Description
Default task restrictions No Select the task restrictions that should be applied to tasks created by this block.

Identification Blocks

We’ve created two Identification blocks to cover both our machine Identification and Identification Supervision capabilities.

Machine Identification

With Machine Identification, you can automate the identification of fields and tables in your submissions.

Manual Identification

Manual Identification allows your keyers to complete Field ID Supervision or Table ID Supervision tasks.

Transcription Blocks

Just as we did with Classification and Identification, we’ve divided our Transcription capabilities into Machine Transcription and Manual Transcription Blocks.

Machine Transcription

In the Machine Transcription Block of your flow, Hyperscience automatically transcribes the content of your submissions.

Manual Transcription

Manual Transcription allows your keyers to manually enter the text found in fields or tables.

Flexible Extraction Block

Depending on your flow’s configuration, Flexible Extraction tasks allow your keyers to validate transcriptions, or add transcriptions to manually categorized Structured pages, which did not go through regular Transcription Supervision.

Collation Block

We’ve created a Collation Block to allow the grouping of files, documents, and pages into cases.

Settings

Name Required? Description
Replace case data from duplicate file names No Replaces case data from repeated file names within the same case.

Complete Blocks

Every flow needs a Complete Block. This block initiates Quality Assurance tasks and changes the submission’s status to “Complete.”

Output Blocks

Output Blocks allow you to send data extracted by Hyperscience to other systems for downstream processing.

Settings shared by all block types

Name Required? Description
Display Name Yes The block's name in Flow Studio.
Description No The block's description in Flow Studio.