[OPENNLP-327] Doccats bag of word feature generator should not use numbers as features

• Type: Improvement
• Status: Closed
• Priority: Minor
• Resolution: Delivered
• Component/s: Doccat
• Assignee: Jörn Kottmann
• Reporter: Jörn Kottmann
• Created: 17/Oct/11 08:48
• Updated: 22/Dec/23 10:24
• Resolved: 22/Dec/23 10:24

Description

It turned out that Doccats bag of word feature generator can be very sensitive to numbers when used for language identification. Therefore numbers should not be included in the bag of words.