Yamaguchi Laboratory, Keio University

Japanese

OSM Lecture & Exercise

Members

Professor

Takahira YAMAGUCHI

Secretary

Yuko MURAKAWA

Assistant Professor

Takeshi MORITA

Research Topics

Graduate Students (Doctor Candidate)

Yun Ki HONG Masao OKABE Hironori TAKEUCHI

Graduate Students (Master Course 2nd Grade)

Yasuhiro AIHARA Naoki FUJII Chie IIJIMA Susumu TAMAGAWA
Koji MIYAHARA Ryuzo YAMAMOTO

Graduate Students (Master Course 1st Grade)

Tatsuya ISHIKAWA Yoshitaro ENOMOTO Keishun OH Shotaro KOBAYASHI
Kenta SUZUKI Yudai SUZUKI Muneaki HOSHINA

Undergraduate Students

Airi GOTO Hiroaki KANEDA Yuka SEKIMOTO Yusuke TAGAWA Shinya MATSUI

Top

Research

DODDLE-OWL: Interactive Domain Ontology Development with Open Source Software in Java (2004-2008)

Abstract. In this paper, we propose an interactive domain ontology development environment called DODDLE-OWL.DODDLE-OWL refers to existing ontologies and supports the semi-automatic construction of taxonomic and other relationships in domain ontologies from documents. Integrating several modules, DODDLE-OWL is a practical and interactive domain ontology development environment.

In order to evaluate the efficiency of DODDLE-OWL, we compared DODDLE-OWL with popular manual-building method. In order to evaluate the scalability of DODDLE-OWL, we constructed a large sized ontology over 34,000 concepts in the field of rocket operation using DODDLE-OWL. Through the above evaluation, we confirmed the efficiency and the scalability of DODDLE-OWL. Currently, DODDLE-OWL is open source oftware in Java and has 100 and more users from 20 and more countries.

Learning a Large Scale of Ontology from JapaneseWikipedia (2008-2010)

Abstract. Here is discussed how to learn a large scale of ontology from Japanese Wikipedia. The learned ontology includes the following properties: rdfs:subClassOf (IS-A relationships), rdf:type (class-instance relationships), owl:Object/DatatypeProperty (Infobox triples), rdfs:domain (property domains), and skos:altLabel (synonyms). Experimetal case studies show us that the learned Japanese Wikipedia Ontology goes better than already existing general linguistic ontologies, such as EDR and JapaneseWordNet, from the points of building costs and structure information richness.

Support for Externalization of Intelligence Skill Using Ontology and Rule-based System (2006-2010)

Abstract. In this paper, we propose the method of externalizing expert and manager's intelligence skill for the knowledge transfer with a domain ontology, a rule ontology and a rule-based system. Specifically we suggest the maintenance support of the rule-based system using a domain ontology to externalize expert's skill, and the support of materialization of business objective using a rule ontology to take manager's ideas out and reflect it in the site. Our proposal was experimentally applied to some actual job at a remote control and maintenance office of hydroelectric power stations in Tokyo Electric Power Co., Inc. We also did evaluation experiments for this case study and have confirmed that our proposal is effective.

Human Robot Interaction Based on Wikipedia Ontology and Robot Action Ontology (2009-2010)

Abstract. This paper discusses how to develop Human Robot Interaction (HRI) based on Wikipedia Ontology and Robot Action Ontology. For the purpose, we implement a system which enables robots to communicate with users in various topics with voice and perform related actions to the topic using a full programmable humanoid robot named Nao. In this study, Wikipedia Ontology, which is developed by extracting the general concepts and relationships among them from Japanese Wikipedia, is utilized as a dictionary for natural language understanding. Robot Action Ontology is also developed for robot action control. We implement the system by relating these two ontologies. From the view point of development, however, there is a wide difference between these ontologies. Therefore, we propose the efficient method for relating these ontologies manually using rdfs:label. Case studies show us how HRI goes well with two topics: healthcare and movie directors.


Top

Copyright (c) Yamaguchi Laboratory, Keio University