WH/T 90-2020 in English

VALID

Unity description for Chinese character identification

Issued on:2020-09-01
Implemented on:2021-01-01
File Format:PDF
Delivery:Via email within 1~3 business days

Price(USD)： $150.00
$146.00

Standard No:	WH/T 90-2020
Document status:	VALID
Title in English:	Unity description for Chinese character identification
Title in Chinese:	汉文古籍文字认同描述规范
Language:	English
File Format:	Electronic (PDF)
Delivery:	Via email within 1~3 business days
Issued on:	2020-09-01
Implemented on:	2021-01-01
ICS Classification:	01.140-Information sciences. Publishing
Chinese Classification:	A14-Library, Archives, Literature and Information
Professional Classification:	WH-Culture
Related Keywords:	general standard chinese character table standard technological breakthrough character character identity private character chinese character identification introduction analysis
Related Topics:	Word ancient books text and data recognize describe the result word description How to describe in the paper Test English description English description

本标准规定了汉文古籍文字认同描述的元数据、文字认同规则描述以及文字认同实例描述的内容、结构及各要素的描述规则。
本标准适用于图书馆及相关机构开展汉文古籍数字化工作中对文字认同过程和结果进行描述。民国时期文献的文字认同可参考执行。

Introduction

Analysis of the core content of the standard

Element category	Traditional processing method	Requirements of this standard	Technological breakthrough
Character set specification	Use private character set	Mandatory use of Unicode basic set	Achieve cross-system data exchange
Variant character processing	Manual experience judgment	Based on the "General Standard Chinese Character Table" and other standards	Establish traceable conversion rules
Metadata records	Unstructured records	8 major categories of standard metadata fields	Support machine learning training

Key technical points

The standard innovatively proposes a three-factor model for character identity: character shape features (structure, strokes), character pronunciation association (ancient and modern sound changes), and semantic context (contextual relationship). In the Yongle Encyclopedia digitization project, this model was used to successfully solve the standardization of 1,243 difficult variant characters.

Implementation Suggestions

Establish a project-level character identity rule library and regularly maintain the version update mechanism
For large series such as the Siku Quanshu, it is recommended to formulate differentiated rules according to the division of classics, history, and collections
Introduce IDS (Ideographic Description Sequence) to describe characters outside the collection

Historical Document Processing Case

In the digitization of the Song Dynasty edition of Wenxuan, it was found that the character "峕" has three variants:
1. 峕 (U+5CD5)
2. 𡵂 (U+21D42 in the extended area)
3. 时 (vulgar writing)
According to clause 4.3.2 of the standard, it is uniformly recognized as "时" (U+65F6), and the original glyph features are recorded in the metadata.