- Print
- DarkLight
- PDF
The Data Extraction tab helps you create rules that you can use in different Ushurs. Only after these rules are created is when you can use the Data Extraction module within an Ushur to extract data.
While creating a Data Extraction module for the Ushur, the list in the dropdown called Extraction Method is based on the rules created in the Data Extraction tab.
Having received an email from an end user, you can extract important information such as user identification and the user’s request using Data Extraction. The tab helps you create rules you define.
Two forms of data extraction rules can be created:
Regex or regular expressions
LI Engine or Ushur’s Language Intelligence
Create a Regex Data Extraction Rule
Regex Data Extraction rules are based on predefined regular expressions. The data that matches the regular expression will be extracted. The format from the end user will have to be an exact match as defined in this rule.
To create a Regex Data Extraction rule, complete the following:
Click the Data Extraction tab on the top part of the dashboard.
This displays the Data Extraction page.Click the Create Rule button.
This displays the Create Rule dialog box.In the Name box, enter the name for the rule.
In the Description box, enter a brief description of the rule.
Ensure that under the Type option, Regex is selected.
In the Regular expression box, add the regular expression you want to be extracted.
In the ROI box, enter the static piece of text that precedes the value or the text pattern you are searching for.
Note
There could be times when similar data types exist in multiple places within a communication.
Click Save.
Create a LI Engine Data Extraction Rule
LI Engine rules are based on Language Intelligence. LI lets you add multiple formats and options.
For instance, if you want to extract date information, you can select Date from the Value Type dropdown. You can then add different phrases or formats on how your end user may add the date. This will help extract data from the end user though they may use different formats for the data.
Note
Sometimes, the system might take time to learn. Therefore, ensure that you get the right kind of phrases. Update the rules based on your standard end-user behavior and update the list.
To create a LI Engine data extraction rule, complete the following:
Click the Data Extraction tab on the top part of the dashboard.
This displays the Data Extraction page.Click the Create Rule button.
This displays the Create Rule dialog box.In the Name box, enter the name for the rule.
In the Description box, enter a brief description of the rule.
Ensure that under the Type option, LI Engine is selected.
This displays the LI Engine option.Under the Value Type dropdown, select the value you want to extract.
In the Extraction Section box, add the content you want to search for.
Click Phrases; this expands the dialog box and displays the phrases you can add for this rule.
Note
You can add as many phrases to a particular rule as you like. The more phrases that you add, often, the better the machine learns.
In the Add Phrases box, enter the phrase.
To add other phrases, click the Plus icon next to Add Phrases.
Click Save.