A study on consistency of name space for utilizing open data

• Project/Area Number: 16K00421
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Web informatics, Service informatics
• Research Institution: Kagoshima University
• Principal Investigator: Fuchida Takayasu 鹿児島大学, 理工学域工学系, 准教授 (70253911)
• Project Period (FY): 2016-10-21 – 2019-03-31
• Project Status: Completed (Fiscal Year 2018)
• Budget Amount: ¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000); Fiscal Year 2018: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000); Fiscal Year 2017: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000); Fiscal Year 2016: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
• Keywords: オープンデータ / 述語ベクトル / RDF / CSV / 連携可能性 / コサイン類似度 / LOD / Word2Vec / 述語ベクトル法 / 名前空間

Outline of Final Research Achievements

An open data policy is being promoted to disclose data held by national and local governments and make them widely available in society, but in reality it is used because the format and content of open data published to each locality is different There are a lot of situations that are not present.
In this research, in order to evaluate the linkability of open data in this country, focusing on predicates in RDF format, generate predicate vectors from the data of CSV files published as open data, and calculate the similarity between open data Proposed a method to evaluate in
We experimented with about 10,000 open data of the whole country and showed that the similarity between open data can be expressed numerically to some extent.

Academic Significance and Societal Importance of the Research Achievements

国の方針としてオープンデータ推進政策がすすめられているが、現状では地方公共団体が公開しているオープンデータが活用されていないのが問題となっている。この問題はオープンデータの公開形式がバラバラで統一性がないことが原因の一つであると考えられる。
本研究ではこの問題を解決する方法として述語ベクトル法と呼ぶ手法を提案した。この手法はオープンデータのデータそのものに着目し、各列のデータが何を表しているかをベクトルの形で表現するものである。本手法を用いてオープンデータ間の類似度を算出し、数値でオープンデータの連携化の姓を示すことが可能となることを示した。

Research Products

• [Presentation] A comparison between Word2Vec and FastText regarding with suggestion of predicates of OpenData (2019)
  • Author(s): Bo Chen, Tadanori Hisanaga, Daisuke Noto, Daiki Tomari, Takayasu Fuchida
  • Organizer: The Twenty-Fourth International Symposium on Artificial Life and Robotics
  • Int'l Joint Research
• [Presentation] Calculation of cooperation possibility of open data using predicate vector method (2019)
  • Author(s): Tadanori Hisanaga, Bo Chen, Daisuke Noto, Daiki Tomari, Takayasu Fuchida
  • Organizer: The Twenty-Fourth International Symposium on Artificial Life and Robotics
  • Int'l Joint Research
• [Presentation] A proposal of recommendation method of predicates for Open Data using statistical method (2018)
  • Author(s): Tadanori Hisanaga
  • Organizer: Symposium on Artificial Life and Robotics 2018
  • Int'l Joint Research
• [Presentation] A study on learning method of Word2Vec for recommending predicates of Open Data (2018)
  • Author(s): Bo Chen
  • Organizer: Symposium on Artificial Life and Robotics 2018
  • Int'l Joint Research
• [Presentation] A proposal of recommendation method of predicates for Open Data using statistical method (2018)
  • Author(s): Tadanori Hisanaga
  • Organizer: Asia-Pacific Conference on Engineering & Natural Sciences
  • Int'l Joint Research
• [Presentation] オープンデータの活用に関する研究 (2017)
  • Author(s): 久永忠範
  • Organizer: 火の国情報シンポジウム2017
  • Place of Presentation: 鹿児島大学工学部（鹿児島県鹿児島市）
  • Year and Date: 2017-03-01
• [Presentation] A proposal of a method for converting into RDF in Open Data (2017)
  • Author(s): Tadanori Hisanaga
  • Organizer: 22nd International Symposium on Artificial Life and Robotics 2017
  • Place of Presentation: B-Con Plaza （大分県別府市）
  • Year and Date: 2017-01-19
  • Int'l Joint Research
• [Presentation] オープンデータのRDF変換の研究 (2017)
  • Author(s): 久永忠範
  • Organizer: 第70回電気情報関係学会九州支部連合大会
• [Presentation] 自治体オープンデータ向けの単語ベクトルの学習法の研究 (2017)
  • Author(s): 陳博
  • Organizer: 第70回電気情報関係学会九州支部連合大会