GB/T 32912-2016 in English

As the basic specification for the information processing of traditional Mongolian, this standard mainly includes two core modules:

• Traditional Mongolian word list (including 6 dimensions such as word form, Latin transcription, and part of speech)
• Formulation additional component list (including 5 dimensions such as grammatical marker code)

Deformed characters: In traditional Mongolian, the shape variants produced according to the change of character position, such as the different writing forms of the beginning, middle and end of a word.

Implementation case: In the development of Mongolian input method, it is necessary to implement automatic character deformation processing according to this standard.

With the acceleration of Mongolian informatization, the original standard has the following problems:

1. Lack of unified norms for word form processing
2. Imperfect system for annotation of structural components
3. Inadequate connection with Unicode encoding

This standard establishes a standardized processing framework by integrating authoritative materials such as the Mongolian Orthographic Dictionary.

Software developers should focus on:

• The mapping relationship between vocabulary data and Unicode encoding
• The implementation of dynamic stem processing algorithm
• The rules of male and female word form changes

Typical application: The Mongolian OCR system developed by Inner Mongolia University has achieved a recognition accuracy rate of 98% based on this standard.