config fastner

Token based rules

Each rule consists a list of elements, specifying the tokens to be matched. Token based rules support 4 wildcards:

1. \w+: represent any token, including a word or punctuation
2. > followed by a number X: represent any number greater than X
3. \< followed by a number X: represent any number smaller than X
4. \d+: reprsent any number.

Examples:

• \w+ cough This rule will match "dry cough", "productive cough", etc.
• temp \> 38 \<40 This rule will match "temp 38.1", "temp 39.5", etc.

Additionally \( and \) can be used to match a subset of a rule, for example:

• dry ( cough This rule will match the word "cough" in the phrase of "dry cough"

Character based rules

The FastCNER support character based rules, which support following wildcards: ( Beginning of capturing a group ) End of capturing a group

\ plus following characters

• p A punctuation
• • An addition symbol (to distinguish the "+" after a wildcard)
• ( A left parentheses symbol
• ) A right parentheses symbol
• d A digit
• C A capital letter
• c A lowercase letter
• s A whitespace
• a A Non-whitespace character
• u A uncommon character: not a letter, not a number, not a punctuation, not a whitespace
• n A return