Teaching Computers to Read Korean: Big Data and Artificial Intelligence at Adan Mun'go

Document Type

Article

Source

Muncha wa sagang

Publication Date

10-2018

Volume

3

First Page

107

Last Page

115

Abstract

The Korean Alphabet is widely praised for it's simplicity. It can be learned by a person in a matter of days. Despite being easy for people o learn, the Korean alphabet, and Korean orthographies more generally, remain exceedingly difficult for computational systems "to learn" despite advances in machine learning and artificial intelligence. The richness of the archive at Adan Mun'go presents an opportunity to teach computers "to read": Korean, especially older documents from the early part of the twentieth century and before.

In this short essay, I hope to entertain you with how the archive at Adan Mun'go is helping to teach computers to read Korean. I will also try to suggest why this work is essential if we wish to make use of new technologies to investigate Korea's rich textural and cultural history.

-article excerpt-

Comments

Also found under the alternative title: "The Future of Adan Mun'go's digital Archive: Big Data and Artificial Intelligence at Adan Mun'go"

Share

COinS