A Method of Reasoning and Learning for Various Data on the Semantic Web

2022 Fiscal Year Final Research Report

• Project/Area Number: 18K11547
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 62020:Web informatics and service informatics-related
• Research Institution: The University of Electro-Communications
• Principal Investigator: Kaneiwa Ken 電気通信大学, 大学院情報理工学研究科, 教授 (00342626)
• Project Period (FY): 2018-04-01 – 2023-03-31
• Keywords: セマンティックウェブ / RDF

Outline of Final Research Achievements

We have proposed two methods of machine learning for ontology and graph data on the Semantic Web. The first method enables us to infer concept subsumption and entity relations from ontology axioms. The second method can train the structural features of graph data for node classification tasks. We have evaluated that our methods outperform the accuracies of conventional methods. For the preprocessing of machine learning, we have improved a fast RDF store system that extracts features from ontology and graph data.

Free Research Field

知識表現と推論

Academic Significance and Societal Importance of the Research Achievements

ここ数年で画像認識を中心に機械学習の成果が大きく社会で応用され、続いて人が解釈して入力したWeb規模の事実データがもたらす機械学習の成果に期待が膨らむ。そのためセマンティックWebの膨大なデータを検索するだけでなく、その知識から自動的に推論する機械学習メカニズムを提案した本研究の成果がその技術確立に寄与する。