Add the following to any RegEx rule to generically avoid matching to correctly formed domains and URLs.

(?![^\s\.]*\.\w)(?<!\.[^\s\.]{0,999})

• Tests on List of Doctor Who universe creatures and aliens
• All times from the fastest on repeat testing

On '\b' start and end of every word for high numerical occurrence

• \b(?![^\s\.]*\.\w)(?<!\.[^\s\.]*) 491ms
• \b(?<!\.[^\s\.]*)(?![^\s\.]*\.\w) 493ms //Somewhat unexpected given the results for 'a' words below.
• \b(?<!\.[^\s\.]*) 408ms
• \b(?![^\s\.]*\.\w) 385ms
• \b 319ms

Word with 'a' on boundary

• \ba(?![^\s\.]*\.\w)(?<!\.[^\s\.]*) 24ms
• \ba(?<!\.[^\s\.]*)(?![^\s\.]*\.\w) 22ms
• \ba(?![^\s\.]*\.\w) 20ms
• \ba(?<!\.[^\s\.]*) 18ms
• \ba 17ms

Realistic test, 'state' occurs six times in article during test.

• \bstate\b(?![^\s\.]*\.\w)(?<!\.[^\s\.]*) 2ms
• \bstate\b(?<!\.[^\s\.]*)(?![^\s\.]*\.\w) 2ms
• \bstate\b(?<!\.[^\s\.]*) 2ms
• \bstate\b(?![^\s\.]*\.\w) 2ms
• \bstate\b 2ms