[OPENNLP-1226] Training an NER model for dates with 'dd.mm.yyyy' as Date format

• Type: Question
• Status: Closed
• Priority: Minor
• Resolution: Fixed
• Fix Version/s: 2.5.0
• Component/s: Name Finder
• Labels: newbie
• Flags: Important
• Assignee: Martin Wiesner
• Reporter: Olga
• Created: 07/Dec/18 10:59
• Updated: 05/Nov/24 09:21
• Resolved: 05/Nov/24 09:21

Description

My txt file for model training has date tags in <START:date> dd.mm.yyyy <END> format. But when I try to use the trained .bin file, the dates are not extracted as they should. My txt tagged file is written one sentence in line. I was wondering maybe the format, and the fullstops in this date format make a difficulty for the model to learn. In the official OpenNLP documentation I can see there is a bin file with date extraction, but I can't see the txt file containing the tags.

I tried to open this bin as a txt format but I read in Stack Overflow that I can't do that.

https://stackoverflow.com/questions/26140492/how-can-i-view-the-content-of-a-bin-file-in-opennlp