Research of text mining for historical documents and visualization of analysis result

• Project/Area Number: 26730169
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Library and information science/Humanistic social informatics
• Research Institution: Onomichi City University (2015-2016); Ritsumeikan University (2014)
• Principal Investigator: Kimura Fuminori 尾道市立大学, 経済情報学部, 講師 (70516690)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Project Status: Completed (Fiscal Year 2016)
• Budget Amount: ¥3,250,000 (Direct Cost: ¥2,500,000、Indirect Cost: ¥750,000); Fiscal Year 2016: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000); Fiscal Year 2015: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000); Fiscal Year 2014: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
• Keywords: テキストマイニング / 古典史料 / 重要語抽出

Outline of Final Research Achievements

I have to realize the methods in order to achieve research of information extraction and visualization from ancient documents.
I proposed the method using support vector machine in order to extract important terms from historical documents, and constructed its system. I also constructed the system to visualize relationships between persons. This system can be used dynamically in order to support analyzing extracted personal relationships. I define that having a strong personal relationship means that the persons have a similar relationship with the places because we believe that one’s personal status is strongly related with the places they frequent. This method can estimate the relationships among persons who do not necessarily have direct relationships.

Research Products

• [Presentation] Personal Name Extraction from Japanese Historical Documents Using Machine Learning (2015)
  • Author(s): Noriyoshi Nagai, Fuminori Kimura, Akira Maeda, Ryo Akama
  • Organizer: The 5th International Conference on Culture and Computing (Culture and Computing 2015)
  • Place of Presentation: Kyoto, Japan
  • Year and Date: 2015-10-18
  • Int'l Joint Research
• [Presentation] Extracting Key Phrases for Suggesting Annotation Candidates from Japanese Historical Document (2015)
  • Author(s): Takafumi Sato, Makoto Goto, Fuminori Kimura, and Akira Maeda
  • Organizer: Digital Humanities 2015
  • Place of Presentation: Sydney, Australia
  • Year and Date: 2015-07-01
  • Int'l Joint Research
• [Presentation] Developing a Collaborative Annotation System for Historical Documents by Multiple Humanities Researchers (2014)
  • Author(s): Takafumi Sato, Makoto Goto, Fuminori Kimura, and Akira Maeda
  • Organizer: The 7th International Conference on Computer Science and Information Technology (ICCSIT2014)
  • Place of Presentation: Barcelona, Spain
  • Year and Date: 2014-12-23
• [Presentation] 東大寺要録からの歴史知識情報の抽出 ―注釈情報の共有を目指して― (2014)
  • Author(s): 佐藤 貴文, 後藤 真, 木村 文則, 前田 亮
  • Organizer: 人文科学とコンピュータシンポジウム
  • Place of Presentation: 国立情報学研究所（東京都）
  • Year and Date: 2014-12-13
• [Presentation] 役者評判記からの人物表現抽出手法の提案 (2014)
  • Author(s): 永井 規善, 前田 亮, 木村 文則, 赤間 亮
  • Organizer: 人文科学とコンピュータシンポジウム
  • Place of Presentation: 国立情報学研究所（東京都）
  • Year and Date: 2014-12-13
• [Presentation] Method for Supporting Analysis of Personal Relationships through Place Names Extracted from Documents (2014)
  • Author(s): Fuminori Kimura and Akira Maeda
  • Organizer: Digital Libraries 2014: ACM/IEEE Joint Conference on Digital Libraries (JCDL 2014) and International Conference on Theory and Practice of Digital Libraries (TPDL 2014)
  • Place of Presentation: London, U.K
  • Year and Date: 2014-09-10