COEセミナーのお知らせ
We would like to inform you about COE-SEMINAR as follows.
DATE : July 26, 2007 15:00-17:00
PLACE : IS Collaboration room 7 (IS Ⅲ-5)
SPEAKER and SUBJECT
: Prof. Lori A. Clarke and Prof. Leon J. Osterweil, (clarke | ljo @cs.umass.
edu) Laboratory for Advanced Software Engineering Research (LASER)
Department of Computer Science University of Massachusetts Amherst, MA
01003 USA
"Precise Definition of Scientific Processes and the Provenance of Scientific
Datasets"
: Prof. Koichiro Ochimizu, School of Information Science, JAIST
"Definition and Realization of Software Accountability"
SPEECH : English
Reference: Research Center for trustworthy e-Society (E-mail: coe-trust)
--------------------------
平成１９年７月２３日
関係各位
２１世紀COEプログラム
－検証進化可能電子社会－
拠点リーダー 片 山 卓 也
ＣＯＥセミナーの開催について
７月２６日(木）、下記のとおりＣＯＥセミナーを開催致します。
つきましては、多数の皆様にご参加いただきたく、ここにセミナーの案内を申し
上げます。本セミナーに関しましては参加申し込み・予約等は必要ございませんので
直接、会場にお越しくださいますようよろしくお願い致します。
記
日時 平成１９年７月２６日(木） １５：００～１７：００
場所 情報科学研究科研究棟 Ⅲ棟５Ｆコラボレーションルーム７
講演者
L.A. Clarke, L.J.Osterweil マサチューセッツ大学計算機学科教授
先進ソフトウェア工学研究センター
講演タイトル
科学プロセスの正確な定義と科学データの出所
講演者
落水浩一郎 情報科学研究科教授
講演タイトル
ソフトウェアアカウンタビリティの定義と実現
講演要旨
Prof. L.A. Clarke, Prof. L.J.Osterweil With the availability of powerful
computational and communication capabilities, scientists now readily access
large, complicated datasets derived by other scientists and build on those
results to produce, through further processing, yet other derived datasets
of interest to themselves and others. For scientists to evaluate the
soundness of these datasets, to reproduce these results, and to build upon
them in responsible and appropriate ways, it is necessary that their
provenance, namely the history of how they were derived, to be documented
precisely and completely. This paper introduces the analytic web, a
synthesis of two types of graphs whose purpose is to provide precise
specification of the scientific processes employed in building scientific
datasets, and to use those definitions to define annotations that could be
attached to the datasets themselves. The work on scientific process
definition that is described here is similar to work often referred to as
"scientific workflow", but emphasizes the use of a semantically richer, more
rigorously defined process definition language that was first developed to
define software engineering processes. The use of this process definition
approach leads naturally into our approach to defining dataset provenance
documentation. The talk will illustrate the information that might comprise
an analytic web for an example process that ecologists use to measure and
analyze the flow of water through a forested watershed. This is a demanding
scientific process that demonstrates the benefits of using a rich semantic
model, as well as the benefits of having an executable process language that
supports automatic creation of process provenance data, reproducibility, and
analysis for scientific soundness.
落水浩一郎 教授
In this presentation, we propose the new concept "Software Accountability".
Our daily life heavily depends on various kinds of e-society systems.
Therefore the system should be designed, implemented, operated and
maintained to assure us that e-society systems are dependable and
trustworthy.
Software accountability is one of such requirements for Trustworthy
e-Society.
We try to define the concept, software accountability, based on the research
results both in Software Engineering field and in Legal Theory.
In the presentation, we show the definition of software accountability,
software accountability functions, software accountability tree, software
accountability module in turn and then a realization mechanism of software
accountability functions, presenting the architecture to combine the
software accountability module with the existing information system.
We also show the result of a case study by developing the course management
system with software accountability functions that shows the feasibility of
our approach.
問合せ先：安心電子社会研究センター （内線：1261・E-mail: coe-trust)
DATE : July 26, 2007 15:00-17:00
PLACE : IS Collaboration room 7 (IS Ⅲ-5)
SPEAKER and SUBJECT
: Prof. Lori A. Clarke and Prof. Leon J. Osterweil, (clarke | ljo @cs.umass.
edu) Laboratory for Advanced Software Engineering Research (LASER)
Department of Computer Science University of Massachusetts Amherst, MA
01003 USA
"Precise Definition of Scientific Processes and the Provenance of Scientific
Datasets"
: Prof. Koichiro Ochimizu, School of Information Science, JAIST
"Definition and Realization of Software Accountability"
SPEECH : English
Reference: Research Center for trustworthy e-Society (E-mail: coe-trust)
--------------------------
平成１９年７月２３日
関係各位
２１世紀COEプログラム
－検証進化可能電子社会－
拠点リーダー 片 山 卓 也
ＣＯＥセミナーの開催について
７月２６日(木）、下記のとおりＣＯＥセミナーを開催致します。
つきましては、多数の皆様にご参加いただきたく、ここにセミナーの案内を申し
上げます。本セミナーに関しましては参加申し込み・予約等は必要ございませんので
直接、会場にお越しくださいますようよろしくお願い致します。
記
日時 平成１９年７月２６日(木） １５：００～１７：００
場所 情報科学研究科研究棟 Ⅲ棟５Ｆコラボレーションルーム７
講演者
L.A. Clarke, L.J.Osterweil マサチューセッツ大学計算機学科教授
先進ソフトウェア工学研究センター
講演タイトル
科学プロセスの正確な定義と科学データの出所
講演者
落水浩一郎 情報科学研究科教授
講演タイトル
ソフトウェアアカウンタビリティの定義と実現
講演要旨
Prof. L.A. Clarke, Prof. L.J.Osterweil With the availability of powerful
computational and communication capabilities, scientists now readily access
large, complicated datasets derived by other scientists and build on those
results to produce, through further processing, yet other derived datasets
of interest to themselves and others. For scientists to evaluate the
soundness of these datasets, to reproduce these results, and to build upon
them in responsible and appropriate ways, it is necessary that their
provenance, namely the history of how they were derived, to be documented
precisely and completely. This paper introduces the analytic web, a
synthesis of two types of graphs whose purpose is to provide precise
specification of the scientific processes employed in building scientific
datasets, and to use those definitions to define annotations that could be
attached to the datasets themselves. The work on scientific process
definition that is described here is similar to work often referred to as
"scientific workflow", but emphasizes the use of a semantically richer, more
rigorously defined process definition language that was first developed to
define software engineering processes. The use of this process definition
approach leads naturally into our approach to defining dataset provenance
documentation. The talk will illustrate the information that might comprise
an analytic web for an example process that ecologists use to measure and
analyze the flow of water through a forested watershed. This is a demanding
scientific process that demonstrates the benefits of using a rich semantic
model, as well as the benefits of having an executable process language that
supports automatic creation of process provenance data, reproducibility, and
analysis for scientific soundness.
落水浩一郎 教授
In this presentation, we propose the new concept "Software Accountability".
Our daily life heavily depends on various kinds of e-society systems.
Therefore the system should be designed, implemented, operated and
maintained to assure us that e-society systems are dependable and
trustworthy.
Software accountability is one of such requirements for Trustworthy
e-Society.
We try to define the concept, software accountability, based on the research
results both in Software Engineering field and in Legal Theory.
In the presentation, we show the definition of software accountability,
software accountability functions, software accountability tree, software
accountability module in turn and then a realization mechanism of software
accountability functions, presenting the architecture to combine the
software accountability module with the existing information system.
We also show the result of a case study by developing the course management
system with software accountability functions that shows the feasibility of
our approach.
問合せ先：安心電子社会研究センター （内線：1261・E-mail: coe-trust)
