A Big Data Analysis of Yumentingzheng

https://aaron.kr/content/portfolio/word-cloud-big-data-analysis-of-manchu-script-in-weiwenqiju/

Poster presentation at The 50th Fall Comprehensive Conference of the Korea Information and Communication Society (KIICE) (제50회 한국정보통신학회 추계종합학술대회).

The Qing dynasty's Yumentingzheng (御門聽政) is an important document comparable to the Annals of Joseon in Korea. This is a record of conversations between the Qing dynasty emperor and his ministers.

Manchu script is one of the few scripts left in the world that does NOT have an easy way to perform OCR (optical character recognition) on it to scan and read documents into a digital format. There is also no readily available and complete Manchu dictionary available for word lookup. Therefore, pages containing the Manchu script must first be transliterated in a way that computers can perform analysis.

In this study, text that was already transcribed using the Möllendorf method was simplified to the Abkai method (transliterations of Manchu) in order to perform a big data analysis on the text and pull out the most important keywords from this portion of the text, the Weiwenqiji (慰問起居).

Transcript

1. Aaron Daniel Snowberger,
   Choong Ho Lee
   어문청정 빅데이터 분석:
   위문기거 일례
   A Big Data Analysis of
   Yumentingzheng:
   Weiwenqiju
   as an Example
   御
   門
   聽
   政
   慰
   問
   起
   居
   

   View Slide

2. Introduction
   Yumentingzheng(御門聽政), which records the contents of the Qing
   dynasty's discussions with his subjects, is an important document like the
   Annals of Joseon in Korea. This paper describes the methods and steps
   for big data analysis of Yumentingzheng written in the Manchu
   alphabet. In big data analysis of documents written in Manchu characters,
   there are many problems that need to be solved in advance, and research
   on these should be preceded. In this paper, a method of big data analysis
   using the R language was proposed in the stage where the text written in
   Manchurian characters was transliterated into Latin characters through
   a preliminary study to be conducted in the future. In the proposed method,
   the Apkai method was adopted for the transliteration of Yumentingzheng,
   and the results of big data analysis were presented using the text of
   Weiwenqiju(慰問起居).
   

   View Slide

3. Text[1] that was already transcribed by
   the Möllendorf method was converted to
   the Abkai method, and then big data
   analysis was performed with R on the
   frequency of words appearing in the text.
   Möllendorf Abkai
   Möllendorf Abkai Manchu
   0
   1
   š x ᡡ
   0
   2
   c q ᡡ
   0
   3
   ū v ᡡ
   

   View Slide

4. Manchu Text Latin Transliteration
   Manchu Möllendorf Abkai
   The original Manchu
   text was transcribed
   using the Möllendorf
   method.
   The transcription was
   edited to according to
   the Abkai method in
   the previous table.
   The resulting edited
   transcription was used
   in the big data analysis.
   0
   1
   0
   2
   

   View Slide

5. It is necessary to find meaningless
   words among the words displayed in
   large letters to remove in order to
   display the word cloud with
   meaningful and important words.
   1st Wordcloud Analysis
   The words ‘be’ and ‘de’ appear the most often
   because nouns cannot be extracted individually.
   Word Korean English
   0
   1
   be ~을,~를, ~로써, ~로 하여금 particles
   0
   2
   de ~에, ~에서, ~에로, 에
   대하여 at, to
   0
   3
   amban 대신(大臣) Minister
   0
   4
   kemuni 늘, 언제나 always
   0
   5
   aniya 해, 년 year
   0
   6
   yasa 눈(目) eye
   0
   7
   udu ~라 할지라도 even though
   

   View Slide

6. It is necessary to find meaningless
   words among the words displayed in
   large letters to remove in order to
   display the word cloud with
   meaningful and important words.
   2nd Wordcloud Analysis
   Word Korean English
   0
   1
   yasa 눈(目) eye
   0
   2
   amban 대신(大臣) Minister
   0
   3
   kemuni 늘, 언제나 always
   0
   4
   aniya 해, 년 year
   0
   5
   udu ~라 할지라도 even though
   With the particles ‘be’ and ‘de’ removed, as well
   as other pronouns such as ‘bi’, ‘si’, ‘udu’, and
   ‘mini’, we can find the most important words.
   

   View Slide

7. Wordcloud2 Analysis
   The R package wordcloud2 produces a wordcloud with
   the ability to mouseover any word to see its frequency.
   (This requires sorting the table in decreasing order first.)
   Word Count
   yasa 10
   qi 9
   ye 9
   se 9
   ere 8
   ba 7
   amban 7
   aniya 7
   the 7
   kemuni 6
   

   View Slide

8. Conclusion
   This paper presented a big data analysis method of
   literature written in Manchu characters. Since the
   Manchu dictionary package has not been provided in
   the R language until now, a word cloud was created
   with the frequency of words in the current state.
   The method of Romanizing Manchu characters was
   converted to the Apkai method without special
   symbols. In fact, the effectiveness of this method
   was demonstrated by conducting an experiment
   with the Weiwenqiju portion of the Yumentingzheng.
   0
   1
   0
   2
   CREDITS: This presentation template was created by Slidesgo,
   including icons by Flaticon, and infographics & images by Freepik
   

   View Slide

9. [1] Zhuang Jifa, Yumentingzheng, Wenshizhe Press, 2000.
   [2] Manchu alphabet. [Internet] Available:
   https://en.wikipedia.org/wiki/Manchu_alphabet
   [3] Diandian Zhang, Yan Liu, Zhuowei Wang, and Depei Wang, "OCR with the Deep
   CNN Model for Ligature Script-Based Languages like Manchu," Hindawi Scientific
   Programming, vol. 2021, Article ID 5520338, https://doi.org/10.1155/2021/5520338
   [4] Jang Yongsik, Kang Higu, Learning to code in R language, Saengneung Press, 2018.
   [5] Manchu Language. [Internet] Available:
   https://namu.wiki/w/%EB%A7%8C%EC%A3%BC%EC%96%B4
   References
   CREDITS: This presentation template was created by Slidesgo,
   including icons by Flaticon, and infographics & images by Freepik
   

   View Slide