Papers - Miyata, Yosuke

検索能力とその要因との関係モデル

石田栄美， 宮田洋輔， 岩瀬梓

日本図書館情報学会研究大会発表論文集 62, 113-116  2014

日本人の検索能力の測定

岩瀬梓， 南友紀子， 宮田洋輔

三田図書館・情報学会研究大会発表論文集, 9-12  2014

Progress of Research on a New Asteroid Exploration Rover Considering Thermal Control

Y Miyata, T Yoshimitsu, T Kubota

Robot Intelligence Technology and Applications 2, 341-350 (Advances in Intelligent Systems and Computing)  274   341 - 350 2014

ISSN  9783319055817

　View Summary

© Springer International Publishing Switzerland 2014. Asteroids have some clues about the origin of the solar system. In recent years, asteroid exploration by surface explorers has been studied actively. Conventional rovers can move under microgravity environment, but the rovers have not considered thermal control. This paper proposes a novel mobility mechanism for a rover under microgravity environment along with temperature control. The effectiveness of the proposed rover is investigated by simulations and microgravity experiments.

• Access to Document (DOI)

Life span of web pages: A survey of 10 million pages collected in 2001

Teru Agata, Yosuke Miyata, Emi Ishita, Atsushi Ikeuchi, Shuichi Ueda

2014 IEEE/ACM JOINT CONFERENCE ON DIGITAL LIBRARIES (JCDL) (IEEE)     463 - 464 2014

Research paper (international conference proceedings), Joint Work, Accepted

　View Summary

This paper highlights the results of a survival survey and life span study of 10 million web pages, mainly in Japanese, that were collected for NTCIR-3 (web task) in 2001. To calculate web page life span, metadata was collected from Internet Archive's Wayback Machine via Memento. The life span study showed that the average life span of a web page is 1,132.1 days.

• Access to Document (DOI)

ウェブページの寿命: 2001 年に存在した 1000 万ページを対象にした調査

宮田洋輔， 安形輝， 池内淳， 石田栄美， 上田修一

日本図書館情報学会研究大会発表論文集 61, 33-36  2013

レビューサイトにみる集合地の可能性: アマゾンカスタマーレビューを例として

上田修一， 石田栄美， 宮田洋輔

日本図書館情報学会研究大会発表論文集 61, 37-40  2013

Identifying works of Japanese classics for construction of FRBRized OPACs

Takuya Tokita, Maiko Kimura, Yosuke Miyata, Yukio Yokoyama, Shoichi Taniguchi, Shuichi Ueda

Cataloging and Classification Quarterly (Cataloging and Classification Quarterly)  50 ( 5-7 ) 670 - 687 2012.06

Joint Work, Accepted,  ISSN  01639374

　View Summary

A research project was conducted in which proper JAPAN/MARC bibliographic records for 158 major Japanese classical works were identified manually, since existing records contain little information about works included in the resources. This article reports on the detailed method used for work identification, including selecting works, obtaining the bibliographic records to be judged, and building the judgment criteria. The results of the work identification process are reported along with average numbers that indicate the characteristics of certain classics. The necessity of manual identification was justified through an evaluation of searches by author and/or title information in a conventional retrieval system. © Taylor & Francis Group, LLC.

• Access to Document (DOI)

学術論文の構成要素と構造

宮田洋輔， 石田栄美， 池内淳， 安形輝， 上田修一

2012 年度 日本 図書館情報学会春季研究集会発表要綱』.-97  2012

The Extent of the Deep Web in Japanese Institutional Repositories

MIYATA Yosuke, AGATA Teru, IKEUCHI Atsushi, ISHITA Emi, UEDA Shuichi

Journal of Japan Society of Library and Information Science (Japan Society of Library and Information Science)  58 ( 2 ) 97 - 109 2012

Joint Work, Accepted,  ISSN  1344-8668

　View Summary

The more the size of Web increases, the more serious the problem of the deep Web (the Web not accessible to search engines) becomes. McCown et al. (2006) and Hagedorn & Santelli (2008) surveyed extent of deep Web using metadata contained in institutional repositories. In this research, applying the method used in that previous work, we measured the extent of the deep Web on a larger scale using PDF file URLs contained in institutional repositories in Japan in September 2009. The results show that the coverage rate of major search engines (Google, Yahoo! and Bing) is 72%, leaving 28% as the maximum extent of the deep Web. And examination of the characteristics of the files revealed that dynamic URLs and longer URLs are associated with decreased coverage rates for search engines.

• Access to Document (DOI)
• CiNii Research

The Trend of Library and Information Science Research in Japan: A Content Analysis of Research Articles

Marie Sugiuchi, Emiko Habu, Shuichi Ueda, Keiko Kurata, Yosuke Miyata, Masanori Koizumi

LIBRARY AND INFORMATION SCIENCE (MITA SOC LIBRARY INFORMATION SCIENCE)   ( 66 ) 127 - 151 2011.12

Research paper (scientific journal), Joint Work, Accepted,  ISSN  03734447

　View Summary

Purpose: This paper seeks to clarify the research trend of library and information science (US) in Japan from 1970 to 2009.
Methods: Of all the articles published from 1970 to 2009 in Library and Information Science and Journal of Japan Society of Library and Information Science, 826 research articles were examined. For each article, (1) authors, (2) topics, (3) research methods, and (4) the use of theories were analyzed. For (1), the number of authors of each article and the first author's affiliation and occupation were examined. Topics were classified into 14 US topics and these topics were further classified into "library science" and "information science". Empirical research methods were further divided into "data collection method" and "data analysis method". The number of research articles and the transition of the number of pages were also investigated.
Results: Our findings indicate that in the LIS field over the last 40 years in Japan, (1) the number of authors whose affiliation is "university" has increased, (2) the number of "information science" articles has gradually decreased, and the majority of articles are now about "library science", and (3) the proportion of empirical research methods has increased. There were no major changes or distinctive trends in the topics discussed by articles. Due to the adoption of a strict peer-review system in both journals, the number of research articles decreased from the 1990s to 2000s; however, the number of articles has increased in recent years due to the increase in the number of graduate students and the enhancement of research grants in the LIS field.

Detecting academic papers on the web

Ishita E., Agata T., Ikeuchi A., Yosuke M., Ueda S.

Proceedings of the ACM/IEEE Joint Conference on Digital Libraries (Proceedings of the ACM/IEEE Joint Conference on Digital Libraries)     413 - 414 2011.07

ISSN  9781450307444

　View Summary

Our research goal is to develop a search engine for open access to academic papers. English and Japanese test sets were built for detection of academic papers from 20,000 PDF files in each language using five annotators. Six classifiers were trained using similar features for each language. We report F1 of 0.74 for English and 0.54 for Japanese and argue that similar features could easily be generated for other languages as well. © 2011 Authors.

• Access to Document (DOI)

Detecting academic papers on the web

Emi Ishita, Teru Agata, Atsushi Ikeuchi, Yosuke Miyata, Shuichi Ueda

Proceedings of Joint 11th annual international ACM/IEEE joint conference on Digital Libraries(JCDL’11)    413 - 414 2011.06

Joint Work, Accepted

Detecting academic papers on the web

E Ishita, T Agata, A Ikeuchi, M Yosuke, S Ueda

Proceedings of the 11th annual international ACM/IEEE joint conference on …  2011

論文から見た日本の図書館情報学研究の動向

杉内， 真理恵， 羽生， 笑子， 上田， 倉田， 宮田， 洋輔， 小泉， 公乃

Library and information science, 127-151 (Mita Society for Library and Information Science)  66   127 - 151 2011

ISSN  03734447

　View Summary

Purpose: This paper seeks to clarify the research trend of library and information science (LIS) in Japan from 1970 to 2009.Methods: Of all the articles published from 1970 to 2009 in Library and Information Science and Journal of Japan Society of Library and Information Science, 826 research articles were examined. For each article, (1) authors, (2) topics, (3) research methods, and (4) the use of theories were analyzed. For (1), the number of authors of each article and the first author’s affiliation and occupation were examined. Topics were classified into 14 LIS topics and these topics were further classified into “library science” and “information science”. Empirical research methods were further divided into “data collection method” and “data analysis method”. The number of research articles and the transition of the number of pages were also investigated.Results: Our findings indicate that in the LIS field over the last 40 years in Japan, (1) the number of authors whose affiliation is “university” has increased, (2) the number of “information science” articles has gradually decreased, and the majority of articles are now about “library science”, and (3) the proportion of empirical research methods has increased. There were no major changes or distinctive trends in the topics discussed by articles. Due to the adoption of a strict peer-review system in both journals, the number of research articles decreased from the 1990s to 2000s; however, the number of articles has increased in recent years due to the increase in the number of graduate students and the enhancement of research grants in the LIS field.

• Access to Document (DOI)
• CiNii Research

The deep web in institutional repositories in Japan.

Teru Agata, Yosuke Miyata, Atsushi Ikeuchi, Shuichi Ueda

Navigating Streams in an Information Ecosystem - Proceedings of the 73rd ASIS&T Annual Meeting, ASIST 2010, Pittsburgh, PA, USA, October 22-27, 2010 (Proceedings of the ASIST Annual Meeting)  47 2010.11

　View Summary

Investigating the current status of the deep web is important for both the general public and researchers. Several deep web surveys have recently been conducted on institutional repositories (IRs). We calculate the extent of the deep web based on the content of searchable IRs in Japan, but using a more appropriate interval and exhaustive search with three major search engines (Google, Yahoo!, and Bing). The deep web is roughly 30% of major search engines' coverage.

• Access to Document (DOI)

A search engine for Japanese academic papers

Ishita E., Agata T., Ikeuchi A., Michiko N., Yosuke M.

Proceedings of the ACM International Conference on Digital Libraries (Proceedings of the ACM International Conference on Digital Libraries)   2010.08

ISSN  9781450300858

　View Summary

A search engine for Japanese academic papers rendered in PDF is described. Evaluation results indicate fewer zero-result queries and higher precision in the top-10 documents than was obtained for the same Japanese queries using Google Scholar or Scirus.

• Access to Document (DOI)

A search engine for Japanese academic papers

Emi Ishita, Teru Agata, Atsushi Ikeuchi, Nozue Michiko, Yosuke Miyata, Shuichi Ueda

Proceedings of Joint JCDL/ICADL International Digital Libraries Conference(JCDL10)    379 2010.06

Joint Work, Accepted

A search engine for Japanese academic papers

E Ishita, T Agata, A Ikeuchi, N Michiko, M Yosuke, S Ueda

Proceedings of the 10th annual joint conference on Digital libraries, 379-380  2010

Information redundancy effect on watching TV news

Matsukawa Rei, Miyata Yosuke, Ueda Shuichi

Library and Information Science (Mita Society for Library and Information Science)  62   193 - 205 2009.12

ISSN  03734447

　View Summary

Purpose: In recent years, the problems of tabloidization of TV news programs and abuse of telops in TV news have grown. Telops in TV news were originally intended to improve understanding of the contents by introducing an information redundancy effect. This study examined by experiment the information redundancy effect of telops when watching TV news.Methods: The experiment included two components: 1) tracking the eye movements of subjects when watching TV news, and 2) examining the degree to which the subjects understood the news contents. As subjects, 22 undergraduate and graduate students participated in the experiment. The participants were divided into two groups: a telop group who watched TV news with telops, and a no-telop group who watched the news without any telops. In the experiment, the differences of eye movements and understanding of the news contents were measured and analyzed.Results: The results from eye movement tracking showed that eye movements in the telop group tended to react sensitively to the appearance of telops. Also, the analysis of understanding showed that the correct answer rates of the telop group were higher than those of the no-telop group for several of the questions. Thus, information redundancy affected the understanding of TV new contents in the experiment. However, telops induced some misunderstanding of some questions.

• Access to Document (DOI)
• CiNii Research

Information redundancy effect on watching TV news: analysis of eye tracking data and examination of the contents

松川 怜, 宮田 洋輔, 上田 修一

Library and information science (三田図書館・情報学会)   ( 62 ) 193 - 205 2009.12

Research paper (scientific journal), Joint Work, Accepted,  ISSN  03734447

　View Summary

短報【目的】近年，テレビニュースの娯楽化と，テレビニュースにおけるテロップの濫用が指摘されている。テレビニュースにおけるテロップは，本来，情報の重複性による内容理解の向上を意図されている。本研究は，眼球運動の測定と内容に関する確認テストとによって，ニュースの視聴に対するテロップの情報重複性の効果を明らかにすることを目的としている。【方法】ニュースの内容理解に対する情報の重複性の効果を検証するために実験をおこなった。実験は，1)ニュース視聴時の眼球運動の測定と2)ニュースの内容に関する確認テストの2過程からなる。22人の大学生と大学院生とが実験に参加した。実験参加者を，テロップがあるニュースをみるグループと，テロップのないニュースをみるグループに分けた。実験では，参加者は三つのニュースを視聴し，視聴中の眼球運動を記録した。その後，確認テストによって，ニュース内容の理解を測定した。【結果】眼球運動測定の結果から，テロップのあるニュースを見たグループの参加者の眼球の動きは，テロップの出現によく反応していることがわかった。確認テストの結果から，テロップのあるグループの参加者は，いくつかの問題で正解率が有意に向上していた。ここからテレビニュースの内容理解に情報重複性の効果があることがわかった。しかし，いくつかの問題は，テロップによって内容を誤解する可能性があることを示唆した。