Everything here is rule-based: values are substituted into your own approved file, and every finding is proved from the text itself. The same input always gives the same result.
What it has learned here, and teaching it a new type from approved examples.
Board ResolutionTamurakoma & Co., Ltd.
9 fields · learned from 3 examples
Release LetterTAISEI CORPORATION
7 fields · learned from 2 examples
Board ResolutionTAISEI CORPORATION
5 fields · learned from 2 examples
Board ResolutionSanyo Engineering & Construction Branch office
3 fields · learned from 4 examples
Release LetterSanyo Engineering & Construction Branch office
3 fields · learned from 2 examples
Return of Income(20-21) 09.06.21Nippon Signal Company Limited
1 field · learned from 2 examples
BOARD RESOLUTIONNippon Signal Company Limited
9 fields · learned from 6 examples
Release LetterNippon Signal Company Limited
9 fields · learned from 4 examples
Release LetterJGC Corporation
7 fields · learned from 2 examples
BOARD RESOLUTIONJGC Corporation
8 fields · learned from 10 examples
TDS Challans-AYInternational Manpower Development Organization (IM) Japan
2 fields · learned from 2 examples
TDS Challans-AY - - PenaltyInternational Manpower Development Organization (IM) Japan
7 fields · learned from 2 examples
Release LetterInternational Manpower Development Organization (IM) Japan
12 fields · learned from 3 examples
Board ResolutionInternational Manpower Development Organization (IM) Japan
13 fields · learned from 4 examples
Pick two or more approved documents of the same kind that differ only in their details — three notices for different shareholders, say. Comparing them shows which parts are the format and which are the facts. The first one you tick is the base, and every draft is built from that exact file, so drafts keep its letterhead and layout.