In system development, it is an important issue how to reuse modules of existing systems. However, in a situation where systems similar to each other but customized differently are required continually, reuse of the modules becomes more difficult because relations among their functions are complicated. To this problem, we aim to establish a technology for extracting a model of functional structure in design documents of the existing system and utilizing it for development of other new systems. In this paper, we introduce our trial approach to extract the model and also discuss how to use a domain ontology in that process.
生命科学分野では Linked Open Data (LOD)が急速に普及しつつあり,これまで数多くのデ ータベースが LOD 上で公開されている.これらのデータベースを有効に活用するためには,公 開されているデータベースの物理的位置に関わらず,複数のデータベースから自在にデータを切 り取り,組合わせて利用できることが望ましい.LOD 上で公開されるデータベースは SPARQL エンドポイントと呼ばれるクエリ記述言語による検索が可能なウェブ API と共に提供されるこ とが多い.SPARQL クエリ記述言語には,SERVICE 句を用いた複数のエンドポイント検索の仕 組みが用意されているが,この機能を利用するためには,どのエンドポイントにユーザが欲しい データがどのような形式で格納されているかを予め知っておく必要がある.一方,著者らはこれ まで SPARQL クエリの記述支援のため,SPARQL Builder Metadata とよぶ各エンドポイントのデ ータスキーマに関するメタデータを収集してきた.そこで,著者らは,これらのメタデータから クラス間関係をベースとした一つのグラフを構築することで,LOD 上のデータを柔軟に切り取 る手法を提案する.そして,そのプロトタイプとして実装した LOD Surfer API を紹介し,その 利用について議論する.
Knowledge graphs are useful for many artificial intelligence tasks. However, knowledge graphs often have missing facts. To populate knowledge graphs, the graph embedding models map entities and relations in a knowledge graph to a vector space and predict unknown triples by scoring candidates triples. Translation-based models are part of knowledge graph embedding models and they employ the translation-based principle. The principle can efficiently capture the rules of a knowledge graph, however TransE, the original translation-based model, has some problems. To solve them many extensions of TransE have been proposed. In this paper, we discuss such problems and models.
In this paper, we proposed methods that develop knowledge graph using ontology matching. Wikipedia, DBpedia, and other Linked Data resources are almost clustered by systematic ontologies, but some resource does not have ontologies it should be linked. "Structuring Wikipedia" project categorizes Wikipedia resources using Extended Named Entity (ENE). Since, DBpedia resources are based on Wikipedia, we use ENE for categorizing DBpedia resources.
Web上で公開されているLinked Open Dataの中から目的に適したデータセットを探索することは難しい。本研究ではLODデータセットを利用したアプリケーションの開発事例を用いたデータセット探索支援手法を提案する。
本研究では,SPARQLクエリ編集者と編集補助者との匿名化マッチングおよび編集支援を実現する機構の試作について述べる.本試作機構は,SPARQLクエリ編集者のプライバシーを考慮し,SPARQLクエリをオントロジーマッピングの動的な補完機構で生成したsemantic relatednessを用いて匿名化を試みる.
Urban areas have many problems such as homelessness, illegally parked bicycles, and littering. These problems are influenced by various factors and are linked to each other; thus, an understanding of the problem structure is required in order to detect and solve the root problems that generate vicious cycles. Therefore, we propose constructing an urban problem linked open data (LOD) system that would include urban problems' causality. In addition, we propose a method for detecting vicious cycles of urban problems using inferences from the LOD. We first designed a Linked Data schema that represents urban problems' causality. Next, we instantiated actual causes and effects using crowdsourcing, supported with techniques based on natural language processing. In addition, we complemented the constructed LOD by drawing inferences using Semantic Web Rule Language (SWRL) rules. Finally, using SPARQL queries, we detected several root problems that led to vicious cycles, then urban-problem experts evaluated the extracted causal relations.
楽器演奏者は各楽器の特性を把握した上で,楽曲の魅力を最大限引き出すための確かな技術と知識が求められる.特に,適切な演奏技術の獲得において土台となる基礎技術の学習・指導プロセスの確立は重要である.中でもクラシックギターは,演奏に関する知識や学習プロセスの標準化や知識共有がなされておらず,演奏者ごとの技能に大きな差が生まれている.そこで本研究では,知識工学的アプローチとしてギター奏法オントロジーを提案する.筆者らの先行研究で構築したクラシックギターの目的指向知識について考察し,その課題解決に向けた知識の構造化を行う.そして構築したギター奏法オントロジーに対するヒアリング調査から,その拡張性や学習・指導現場への有用性についての考察を通して,ギター演奏の基盤となる可能性を確認する.
This paper proposes Crop Vocabulary(CVO) as a basis of core vocabulary of crop names that becomes the guidelines for data interoperability between agricultural ICT systems on the food chain. Since a single species is treated in different ways, there are many different types of crop names. So, we organize the crop name discriminated by properties such as scienti c name, planting method, edible part and registered cultivar name. CVO is also linked to existing vocabularies issued by Japanese government agency and international organization. It is expected to use in the data format in the agricultural ICT system.
We have been building structured manuals for care processes. It is useful to formalize them in a standard form which is defined by W3C because we can re-use some of the semantic technologies to process the structured manuals. This manuscript reports a trial to RDFize the structured manuals for care processes.
We have been prototyping a regional understanding support system by local food from 2016. In version 3, the mochi map becomes Linked Open Data,This paper discusses our perspective about this system and utilizing data in the future.