[Legacy] [PSIcapture 6] How to Configure Basic Classification

Introduction

How to implement Classification in new or existing document types.

Step-by-step guide

1. From the main PSI:Capture screen, click the Configuration icon, and then add or edit a document type.
2. You will need to move to Step 3, and enable “Support Document Record Types”. Click the ‘Configure’ button.
   
    
   
   
3. In the Document Record Types Configuration window, you can add single records and multiple records.
   
    
   
   
4. Next, proceed to Step 4 of 10, and you will need to add a new profile, and set it for Fixed Length, pages per Document = 1
   
    
   
   
   
   
5. Next you will need to proceed to Step 6, and create the index fields that will acquire the data. Some fields can even be assigned to a particular record type if they do not exist across all types.
   
    
   
   
6. Click on the “Zone” tab in Step 6, then click on “Define Zones”.
   
    
   
   
7. Here in Zone Configuration is where you will define profiles for each record type, or document. You will need to click the page with the green plus in order to create a new profile.
   
    
   
   
8. On your new, blank profile, you will need to assign it a name, which commonly uses a name referencing the document. A description is optional, and you will need to change the Trigger to “Activate on Record Type”, and choose the appropriate record type.
   
    
   
    
   
    
   
   
9. Click on the “Zones” tab and from here, you will draw your OCR, Smart, or OMR zones for this record type. Each zone requires a name.  Once your zones are drawn, click the purple disk/page icon on the toolbar to save and return back to Step 6.
   
    
   
   
   
   
10. You will need to ensure each index field is mapped to the proper OCR zone you have drawn. Set the Zone Definition Action to OCR (Or ICR/OMR, depending on your document) and then make sure the zone is defined in the bottom field.
    
    
     
11. Move on to Step 9 of 10. You will need to add Classification from the left list of modules.
    
    
     
12. Once at the Classification Settings dialog, click the ‘Add’ button to define a new Classification profile. On this profile, you will need to name the profile, and then choose which Record Type it will be associated with. You can also define how many pages the documents need to be, or leave at the default of 1.
    
     
    
     
    
    
13. You can import a template image to help aid in defining your Classification rules. Simply add as many rules as you need, and define if they are positive or negative matches. You can define OCR values to match on, either by manually writing a RegEx code or using the built-in generator based on your text selection.
    
     
    
    Once your rule(s) are defined, you can click the orange lightning bolt icon to test your expressions for a match.
    
     
    
    
14. Once your rules are all added, click the save icon to return back to the Classification Setting screen.
    
    
     
15. Next, click on Classification Options tab. Here you can define the Page Validation Options. Use this to enforce document page counts. In this example we are using Strict Page Count, which will match the amount of pages defined on the Form. Another commonly used option is “No Page Count Validation”, with the option enabled  “Combine non-classified documents with previously classified document”. This will not enforce a page count so if you have variable length document, it can be combined until the next classified document is found.
    
     
    
    
    
    
16. Click save on your options, finish your workflow and save the document type. You can now start a new Capture/Import, and bring in your documents with the document type and move it through the workflow. Your results in the end should be all documents with their proper index data, per record type.