Search: extract-structured-data

Each Pipeline defines the structure of Data fields that akaBot Vision extracts. Description When editing this structure you have two options: Use pre-trained Data fields – AkaBot Vision’s Generic AI engine has been pre-trained to recognize specific Data fields and enables you to start extracting data

[3] Configure Fields for Data Extraction

Customizing Data Extract

Configuring Fields for Data Extraction

="wikigeneratedid" id="HParagraph1" %) Each Pipeline defines the structure of Data fields that akaBot Vision extracts. == **Description** == (% class="wikigeneratedid" %) When editing this structure you have two

to recognize specific Data fields and enables you to start extracting data without any additional training

A basic element in the extraction schema is the data field. However, akaBot Vision enables the capture of even more complex structures like tables. Adding a predefined table field If you are missing

settings. In this tab, you can manage pre-trained data fields and select which of them should be extracted

[4] Capture Custom Table Data

Customizing Data Extract

Capturing Custom Table Data in akaBot Vision

="wikigeneratedid" id="HParagraph1" %) A basic element in the extraction schema is the data field. However, akaBot Vision enables the capture of even more complex structures like tables. == **Adding a predefined table

of them should be extracted. (% style="text-align:center" %) [[image:image-20220421003652-1.png||data

When you select a document for review, it will take you to the Validation screen. In the left panel, you can see the predefined list of Data fields that have been extracted. Press Tab to check

be able to export the data. If you realize that the data field was not extracted correctly, you can make

[3] Validate the Data

Validate the Data

, otherwise, you won’t be able to export the data. If you realize that the data field was not extracted

="font-family:Arial,Helvetica,sans-serif" %)In the left panel, you can see the predefined list of Data fields that have been extracted. (% style="font-family:Arial,Helvetica,sans-serif" %)[[[[image:image

RCA.Activities.ComputerVision.CVTableExtract Description Extracts a table that is visible

On Screen* - Chooses the element to extract. This activity will generate a string variable (Selector

to consider data security requirement before using this property. Display Name (String) - The name

CV Table Extract

CV Table Extract

to organize and structure your code better. E.g: [201135155] CV Table Extract **Output** * **Result

" %) ((( RCA.Activities.ComputerVision.CVTableExtract == **Description** == Extracts a table that is visible on the screen and stores

** == * **Indicate On Screen*** - Chooses the element to extract. This activity will generate a string variable

RCA.Activities.ComputerVision.CVGetText Description Extracts the text from a specified UI element. (*For mandatory) In the body of activity Indicate On Screen* - Chooses the element to extract the text

the activity. Remember to consider data security requirement before using this property. Display Name (String

" %) ((( RCA.Activities.ComputerVision.CVGetText == **Description** == Extracts the text from a specified UI element. [[image:image

the element to extract the text. This activity will generate a string variable (Selector) to specify

the activity. Remember to consider data security requirement before using this property. * **Display Name

After importing the document successfully and the data extraction process is successfully finished, the document status will change to TOREVIEW. To Review Tab You can see documents with this status

Shortcuts When reviewing documents in akaBot Vision, there are multiple ways of moving between data fields

="wikigeneratedid" %) After importing the document successfully and the data extraction process is successfully

with this status in the “To review” tab in the user interface. [[image:image-20220420193327-1.png||data-xwiki

in each field that has been detected incorrectly [[image:image-20220420193327-2.png||data-xwiki-image

on: Built-in checks – we perform Data Integrity checks based on values found on the document. Such checks

the Extraction schema you should be seeing the message for the required fields with no captured value. Table

Customizing Data Extract

– we perform Data Integrity checks based on values found on the document. Such checks could

saving the Extraction schema you should be seeing the message for the required fields with no captured

When you send a document to the pipeline, akaBot Vision will immediately start to extract the data from it. Description During this stage, the document has an importing status. In case something went wrong during the upload stage or later during the importing stage, the document will fail

="wikigeneratedid" %) When you send a document to the pipeline, akaBot Vision will immediately start to extract the data from it. == **Description** == (% class="wikigeneratedid" %) During this stage, the document has an importing status. (% style="text-align:center" %) [[[[image:image-20220420191058-1.png||data-xwiki-image

status). Find that all the extracted data is ok, they can click Confirm button then this document

to the Postponed tab (Switching to Postponed status). * Find that all the extracted data is ok, they can click

) - Check if you want to public the activity. Remember to consider data security requirement before using

to organize and structure your code better. E.g: [3424325] Maximize Window

. **Misc** * **Public (Checkbox)** - Check if you want to public the activity. Remember to consider data

. You can edit the name of the activity to organize and structure your code better. E.g: [3424325