WindowCharsBeforeMatch
The number of UTF-8 characters to include in the matching window when Named Entity Recognition finds a potential match.
Named Entity Recognition uses the configured Regex to run an initial check to find potential matches in your input text. It then creates a match window by selecting characters before and after the potential match, and then uses your grammar to find actual matches in these windows.
You configure the window size by using WindowCharsAfterMatch and WindowCharsBeforeMatch
. Named Entity Recognition sets the start of a match window at the closest word boundary before the configured number of characters.
Type: | Long |
Default: | None |
Required: | No |
Configuration Section: | MyPreFilterTask |
Example: | WindowCharsBeforeMatch=100
|
See Also: | TokenWithPunctuation |