On A Real-Time Control Model for Robust Multi Agent Systems Under Uncertain Environment

不确定环境下鲁棒多智能体系统实时控制模型

• 批准号: 16500079
• 负责人: SHINTANI Toramatsu
• 金额: $ 2.24万
• 依托单位: Nagoya Institute of Technology
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (C)
• 财政年份: 2004
• 资助国家: 日本
• 起止时间: 2004 至 2006
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-16500079/
• 关键词: Multi Agent System; Uncertainty; Paper Recommendation; WWW; Qualitative Reasoning; Educational Support; Auction Protocol; Ubiquitous Environment; 文献共有; オークションメカニズム; 組み合わせオークション; 重み付き制約充足問題; アドホックネットワーク; 強化学習; プロダクションルール生成; ユーザモデル; テキスト分類 /

We have conducted researches on cooperation mechanisms of multiagent systems under uncertain environments for massive-scale and real time knowledge processing. We developed an algorithm to effectively solve complex multiagent cooperation, an algorithm for ad-hoc information sharing algorithm based on a multiagent system, and a knowledge discovery algorithm under uncertain environment.We developed a multiagent cooperation framework under uncertain environments for ubiquitous computing environment. We implemented applications of our cooperation framework in order to evaluate the developed system. The applications are (1) ad-hoc information sharing system, and (2) a real time knowledge processing system based on qualitative simulations, describing below. We realized an advanced web implementation methodology, which can be used to directly edit and make web pages on a web browser, for ubiquitous environments to effectively develop the applications.(1)We propose a new keyword extraction met … More hod for generating a user profile using collected papers without using a large corpus. We assume that a user's interest exists in papers. Our method can extract keywords that can express user's interest in papers that user's interest exists. Our method can be used for enhancing the paper collection and sharing system. User profiles are automatically constructed by using the method. We conducted several experiments to show how effectively our method can extract keywords that represent user's interests.(2)We developed a support method of economic learning system based on qualitative simulation. We employ qualitative simulations because this lets the learners can understand the conceptual principles in economic dynamics. This system supports users through making partial-adjacent relationships/graph model and supports users displaying errors about inconsistency and incorrect rules when there are inconsistency and mistaken relationship in the graph model made by users.Our contributions are (a) a development of an information sharing system for large-scale document databases with uncertainty, (b) a development methodology on the web under uncertain environment, (c) a design and implementation method based on qualitative reasoning for educational support with uncertainty, and (4) a cooperation protocol among uncertain agents. Less

本文研究了不确定环境下多智能体系统的协作机制，以实现大规模、真实的时间知识处理。提出了一种有效解决复杂多智能体协作问题的算法，一种基于多智能体系统的ad-hoc信息共享算法，一种不确定环境下的知识发现算法，提出了一种普适计算环境下不确定环境下的多智能体协作框架。我们实现了我们的合作框架的应用程序，以评估开发的系统。应用是（1）ad-hoc信息共享系统，和（2）基于以下描述的定性模拟的真实的实时知识处理系统。我们实现了一个先进的Web实现方法，它可以用来直接编辑和制作Web浏览器上的网页，无处不在的环境，有效地开发应用程序。(1)We提出了一种新的关键词提取方法 ...更多信息 一种使用所收集的论文而不使用大型语料库来生成用户简档的方法。我们假设用户的兴趣存在于论文中。我们的方法可以提取关键字，可以表达用户的兴趣，用户的兴趣存在的论文。我们的方法可以用来加强论文收集和共享系统。通过使用该方法自动构造用户简档。我们进行了几个实验，以显示如何有效地我们的方法可以提取代表用户的兴趣的关键字。(2)We提出了一种基于定性模拟的经济学习系统支持方法。我们采用定性模拟，因为这让学习者可以理解经济动态中的概念原则。该系统通过建立部分邻接关系/图模型来支持用户，当用户建立的图模型中存在不一致和错误的关系时，支持用户显示不一致和不正确的规则错误。我们的贡献是：（a）开发了一个面向不确定性的大型文档数据库的信息共享系统，（B）一种不确定环境下的Web开发方法，（3）基于定性推理的不确定性教育支持系统的设计与实现方法;（4）不确定性Agent之间的协作协议。少

项目成果

Nurse Scheduling System based on Dynamic Weighted Maximal Constraint Satisfaction Problem

基于动态加权最大约束满意度问题的护士排班系统

• 发表时间: 2005
• 期刊: Trans. of the JSAI Vol.20, No.1
• 作者: Tadachika Ozono;Atsushi Isomura;Takayuki Ito;Tadachika Ozono;Toramatsu Shi tani
• 通讯作者: Toramatsu Shi tani

An E-Learning Support System based on Qualitative/Quantitative Simulations

基于定性/定量模拟的电子学习支持系统

• 发表时间: 2006
• 期刊: International Transactions on Computer Science and Engineering vol.26, no.1
• 作者: K.Inata;T.Miyahara;H.Ueda;K.Takahashi;Tokuro Matsuo
• 通讯作者: Tokuro Matsuo

定性シミュレーションに基づく環境教育支援システム

基于定性模拟的环境教育支持系统

• 发表时间: 2006
• 期刊: パーソナルコンピュータユーザ利用協会論文誌 vol.16, no.3
• 作者: Pat Langley;Sachiyo Arai;Daniel Shapiro;松尾徳朗
• 通讯作者: 松尾徳朗

組み合わせオークションにおける効用比較に基づいた架空名義入札者発見法の提案

组合拍卖中基于效用比较的虚拟投标人发现方法的提出

• 发表时间: 2005
• 期刊: 情報科学技術レターズ
• 作者: 小栗崇志;ラック・ターウォンマット;Hiroaki Imade;Haibo Yu et al.;松尾徳朗
• 通讯作者: 松尾徳朗

A Learning Support Method in Qualitative Simulation-based Economic Education

定性模拟经济教育中的学习支持方法

• 发表时间: 2005
• 期刊: Proc. of the 20th National Conference on Artificial Intelligence (AAAI-2005)
• 作者: 小栗 崇志;津田 耕平;Ruck Thawonmas;Tsunenori Mine et al.;T.Onoda;Tokuro Matsuo
• 通讯作者: Tokuro Matsuo