Get started free
Get started free

Extension Kit for M-Files | Documentation

View Categories

General Settings

Enabled #

Specifies whether the intelligence service is enabled or not.

Suggestions → Suggestion (n) #

Settings used for mapping Azure Document Intelligence metadata suggestions with M-Files properties. This is where we connect all the possible Terms that we get from prebuilt and custom Azure models to our M-Files properties.

Term #

Connects Azure Document Intelligence field to M-Files suggestion term. Detail list of all possible Terms is available in Server-Specific Settings configuration page.

Mode #

  • No mapping (default): No mapping is given.
  • Any Property
  • Specific Property

Object Type or Value List #

Shown when “Any Property” is selected under the “Mode” option.

Property under which term suggestion will be recommended.

Property #

Shown when “Specific Property” is selected under the “Mode” option.

Property under which term suggestion will be recommended.

Advanced Settings #

  • Suggest New Values Using Text-Based Matching: Suggest new and existing values using text-based value detection if value mappings have not been configured. If this setting is disabled, only existing values are suggested using text-based matching.
  • Suggest Existing Values Using Text-Based Matching as Fallback: Suggest existing values using text-based value detection if no match is found in value mappings.
  • Suggest New Values Using Text-Based Matching as Fallback: Suggest new and existing values using text-based value detection if no match is found in value mappings. If you wish to enable this setting, “Suggest Existing Values Using Text-Based Matching as Fallback” must be set to “Yes”.

Maximum Processing Time in Seconds #

The maximum amount of time in seconds that the intelligence service waits for the suggestions to be processed and generated before canceling the operation.

Maximum Number of Suggestions per Term #

The maximum number of metadata suggestions to be provided for a single term. The suggestions are ranked by their confidence score, and the ones with the lowest score are not included in the suggestions if the total number of suggestions is higher than this value.