Use keyword detection

Kofax ReadSoft Entrance can search documents for keywords that you specify in a text file. All of the keywords that are found are saved in a field value separated by commas. You can easily change the keywords without disturbing the document profile, since the keywords are in an external text file.

This can be useful for classifying your documents into sub-categories. You get a list of different keywords found on an document, and you use this list to determine the document type. For example, if credit and note are found, then the document is a credit note. (Note: A document is also classified as a credit note if Interpret finds a negative amount. See The Debit/Credit field for more information.)

To use keyword detection, these steps are necessary. They are recommended for advanced users only.

  1. Create a text file for your keywords.
    • Each word or phrase must be on a separate line.
    • The keywords can contain spaces.
    • Keywords are not case sensitive.
    • Blank lines are allowed in the keyword file.
  2. Save the keyword file. Recommended: Put it in the same folder as Eiglobal.ini.
  3. Decide which field type you want to use keyword detection on. The available field types are listed in the Field types dialog.
  4. Open the Eiglobalextra.ini file used by your country profile.
  5. Find the section that corresponds to the field type you want to use keyword detection on.
  6. In the corresponding section:
    • Create a KeyWordFile setting, and set the value to the name of the keyword file. If the keyword file is not located in the same folder as Eiglobal.ini, then specify the full path.
    • Create a KeyWordSeparator setting, and set the value to the separator character you want. This character will be used to separate values when more than one word from the keyword file is found.
      Note
      Do not use the same separator character that you use for your transaction description.
      Example

      [FieldName]

      KeywordFile= keywords.txt

      KeyWordSeparator=,

  7. Save and close Eiglobalextra.ini.
  8. Create a new single item field profile and give it a name.
  9. In the General settings for the field profile, select the field type that you decided to use.
  10. Type "(1-511)" in the Specification box. The other format settings are irrelevant.
  11. In the Titles settings, ensure that Use titles to search for value is selected.
  12. In the Title box, type Dummy (or any other name). Then click Add.

    This adds a dummy title in the Selected titles box. It is not used anywhere but is necessary to make keyword detection work.

  13. Select Title occurrence sets value in the Treatment settings (Options page).
  14. All of the keywords found on a document can be included in the output. To do this, open the transaction description and change the format of the transaction field—the one with the same name as the field that you used keyword detection on—from N(1) to X(1-511).

    In the case of this specific functionality, all of the keywords found on the document are included in the output, in spite of the Title occurrence sets value setting.

    Note The values that are saved to the field and transferred are taken from the keyword file; the actual interpreted values are not used.