This is the publication list of Dr. Behrouz Homayoun Far.
Copies of most of the papers in PDF format are available on-line for preview only. You can preview the electronic PDF papers using Adobe's free Acrobat Reader 4.x or better.
Please note that due to Copyright Transfer Policy of the representative publishers the electronic PDF papers cannot be printed. If you wish to have an electronic printable version of the PDF files please contact me via e-mail, mentioning your name and affiliation, your contact address, papers' title and publication year in the body of your message.
Go to B.H. Far Home Page
How to view PDF files.

P
U
B
L
I
S
H
E
D

P
A
P
E
R
S

1
9
9
9
 
   Research Doc. 1999-01 Click to download PDF version
Authors(s): Romi Satria Wahono, Behrouz H. Far
Title:
"Distributed Expert System Architecture for Automatic Object-Oriented Software Design"
Transactions/Magazines/Conferences:
The Third Workshop on Electro-Communication and Information, WECI-III, pp. 131-134, Tokyo, Japan, March, 1999.
Abstract
The goal of this project is developing an expert system that aims to help software designers while designing object-oriented software by automating object-oriented design process. Some novel points are : using an interactive user interface, to enable software designer and expert system can communicate with each other in order to determine the best design, without disturbing the creativity of the software designer; making it possible for distributed expert system, to overcome the data sharing and the interoperability problems in groupware. In this paper we present the research project outline and the research approaches.
Original Language: English
 


 
   Research Doc. 1999-02 Click to download PDF version
Authors(s): 張 石峰、陳 慧、Behrouz H. Far、河野 善彌
Title:
"自動設計システムとその特性"
Transactions/Magazines/Conferences:
電子情報通信学会技術研究報告 資料 AI98-86, KBSE98-56, pp. 9-16, Chiba, Japan, March, 1999.
Abstract
この論文は交換接続制御プログラムを自働設計する為のエキスパートシステムに関するものである。人によるこの設計作業を調べ、交換の仕様を表す接続の図に付き、ある種の分割を行う事が設計の知識である事を見出し、これをエキスパートシステムとして作った。このシステムは既に原理試作等も行っている。設計システムあるいはソフトウェアシステムとして各種の特性を採取する事を目的として、良い設計を行なった。自動設計の対象とする交換システムを説明し、設計知識の獲得を説明している。次ぎに、このエキスパートの設計は、先行する試作の経験を元に、小さな段階毎に多くの図面を使い、厳しくチェックする等、極力工程を厳しく統制した。試作後の各種の実績を過去実績と対比する等して紹介している。
This paper reports on a development and the result of the developed system, which is an expert system of a switching software development. A Knowledge Engineer studied the human design work and found that the design was made by dividing the specification chart in a way. The expert system has been built to implement the design knowledge. As the system had been developed for an experimental purpose, the design this time has been improved. In order to gain a good design, various cautions have been taken for getting reliable data. The switching system is explained, and the acquisition method is discussed. Thus purifying the process, it is expected that the results are reliable. Several data are shown contrasting several past similar experiences.
Original Language: Japanese
 


 
   Research Doc. 1999-03 Click to download PDF version
Authors(s): 馬 培金、河野 善彌、陳 慧、Behrouz H. Far
Title:
"電話対応をする高級秘書の知識構造"
Transactions/Magazines/Conferences:
電子情報通信学会技術研究報告 資料 AI98-87, KBSE98-57, pp. 17-24, Chiba, Japan, March, 1999.
Abstract
この論文は知識工学的に電子秘書を作る研究であ。組織の幹部に仕える幹部秘書は、標準化された手順で高度で気の利いたサービスを提供する。これをエキスパートモデルにとり、入出力関係に着目し書面により知識を獲得しつつ,知識構造モデルを求める事をボトムアップに行なった。最下位の対話の知識構造から始め、情報取得、判断、交渉の知識モデルを得た。これらは多階層の状態空間の中に位置づけられ、多階層化,状態の細分化を行なえば、如何様にでも高度化できる。具体的な書式での知識獲得、その高度化、構造化の実際を詳しく説明している。このように得た知識構造モデルはそのままソフトウェア化でき、優れた秘書の行動を再現でき、また、細部構造では認知科学の道具になり得る。
This paper reports on an electronic secretary built in Knowledge Engineering way. High class secretaries for top managers of organizations offer excellent and sophisticated services in a standardized way. They are taken as expert model, and their knowledge is acquired by documents putting emphasis on input and output relationship, and their structural knowledge models have been established through such bottom up researches. The starting point was the bottom level conversational knowledge, information gathering knowledge, decision knowledge, and knowledge model for negotiation. They may be situated in a multi-layer state space, and when the number of layers are increased, or each state is chosen to be small, the expert system may be made more sophisticated. Practical procedures for knowledge acquisition using documents, how to structure the knowledge, and grading up of the structures. thus gained knowledge model may be easily implemented by software, and it behaves as an electronic secretary, and the fine structure may be used also in Cognitive Science field.
Original Language: Japanese
 


 
   Research Doc. 1999-04 Click to download PDF version
Authors(s): Hassan Abolhassani, Hui Chen, Zenya Koono, Behrouz H. Far
Title:
"Software Creation: Structure of Human Design Knowledge and Its Application to Automatic Software Design"
Transactions/Magazines/Conferences:
The Fourth Annual International CSI Computer Conference, CSICC' 98, pp. 66-71, Teheran, Iran, January, 1999.
Abstract
This paper reports researches in Software Creation project aiming at automatic design of software. In the project, Knowledge acquisition as modeling has been taken using a high maturity software organization as the expert model. In it, the hierarchical work process becomes solid and unified. By virtue of the hierarchical work process, the analysis becomes possible to endlessly to reach an elementary human mental operation.
Original Language: English
 


 
   Research Doc. 1999-05

Click to download PDF version

Authors(s): Behrouz H. Far
Title:
"Towards an Agency Theory of an Agent Population"
Transactions/Magazines/Conferences:
The 13th Annual Conference of Japanese Society for Artificial Intelligence, JSAI' 99, pp. 454-455, Tokyo, Japan, June, 1999.
Abstract
Although there are many projects focusing on multi-agent systems, there are only a few focusing on systematic design of large scale multi-agent system. Electronic commerce is an example of such systems. In this project, we introduce a model and devise an implementation for the large scale knowledge sharing using Bayesian networks, game theory and network security. The agent model for the system components blends the traditional expert systems' reasoning engine with a multi-layer knowledge base, communication and documentation engines. Electronic Commerce (EC) is a potential application for such a system. EC is viewed as a society of software agents, such as customer, search, catalog, manufacturer, dealer, delivery and banker agents, distributed over the Internet and interact and negotiate with each other.
Original Language: English
 


 
   Research Doc. 1999-06

Click to download PDF version

Authors(s):Anete Hiromi Hashimoto, Hassan Hajji, Behrouz H. Far
Title:
"A Probabilistic Approach for Student Modeling Task"
Transactions/Magazines/Conferences:
The 13th Annual Conference of Japanese Society for Artificial Intelligence, JSAI' 99, pp. 68-69, Tokyo, Japan, June, 1999.
Abstract
Student modeling is the task of gathering information about the target student in order to offer tailored instruction. To achieve this purpose, the formalism chosen to represent the student module should be able to draw estimations based on a set of restricted and uncertain information as well evolving aspects as a learning process implies. The approach propose is the dynamic probabilistic network which is a formalism suitable for representing evolving information in an uncertain environment. The probabilistic student module will depict the student's cognitive states in a domain independent manner which will be used as the source of information to prepare a general instructional plan. This plan, combined with the student's response diagnose, is refined to a more specific instructional action. In this article, the basics of dynamic probabilistic student module will be presented.
Original Language: English
 


 
   Research Doc. 1999-07

Click to download PDF version

Authors(s):Romi Satria Wahono, Behrouz H. Far
Title:
"OOExpert: Distributed Expert System for Automatic Object-Oriented Software Design"
Transactions/Magazines/Conferences:
The 13th Annual Conference of Japanese Society for Artificial Intelligence, JSAI' 99, pp. 456-457, Tokyo, Japan, June, 1999.
Abstract
Researchers and software designers come to a conclusion that identifying objects including associations, attributes is an ill-defined task and there is no unified methodology for object-oriented software design. In this research, we suggest an approach to solve the difficulties of object-oriented design by developing a distributed expert system that contains formulated design patterns and rules that aims at automating object-oriented design process in an interactive way. A novel point is using an interactive user interface to enable software designer and the system communicate with each other in order to determine the best design, without disturbing the creativity of the designer. Thus, making it possible for distributed expert system, to overcome the data sharing and the interoperability problems in a groupware. In this paper we present the system architecture and the research outlines.
Original Language: English
 


 
   Research Doc. 1999-08

Click to download PDF version

Authors(s):朱 仲武,王 樵,近藤 邦雄,B.H. ファー
Title:
"汎用的なWebベース個人適応型学習支援システム構築ツールWWW-CALISTとその仕様について"
(WWW-CALIST: A General Purpose Tool for Constructing Web-Based Individualized Adaptive CAL Systems),
Transactions/Magazines/Conferences:
教育システム情報学会誌, vol. 16, no. 1, pp. 14-24, April, 1999.
Abstract
It seems that a big problem in utilizing the WWW to education fields is that teachers have difficulty directly developing and managing a WWW-based intelligent CAL (Computer Aided Learning) system, because of the shortage of funds and teachers who are skilled in both AI and computer information technologies. Focusing on solving the problem, we have developed a general-purpose tool called WWW-CALIST (WWW-based Computer Aided Learning Individual-adaptive Systems Tool) , which helps teachers build up a web-based individualized adaptive computer aided learning environment called "test - search information using an adaptive knowledge map -Adaptive learning". The WWW-CALIST has some such functions and features as follows:
  • Through a register window, a teacher can input his teaching strategies into a knowledge base easily. And adaptive knowledge maps guiding an adaptive learning can be created on-line by the system.
  • Through a monitor window, a teacher can immediately supervise the operation situation of the system, and according to the feedback information modify his teaching strategies at any time.
  • By using easy-to-use interfaces equipped with the tool, a teacher who knows only the HTML can develop and manage web-based individualized adaptive CAL systems for various education domains.
  • The tool is developed in JAVA and its upper exchanging language HORB, so it is OS independent.
Original Language: Japanese
 


 
   Research Doc. 1999-09

Click to download PDF version

Authors(s):H. chen, H. Takano, H. Abolhassani, Z. Koono, Behrouz H. Far
Title:
"An Approach to an Integrated Intelligent CASE Tool for Automatic Software Design"
Transactions/Magazines/Conferences:
Proc. 11th International Conference on Software Engineering and Knowledge Engineering, SEKE' 99, pp. 310-314, Kaiserslautern, Germany, June 16-19, 1999.
Abstract
This paper introduces system design of an Integrated Intelligent CASE tool, that may be applicable from top level (data flow diagram) design down to coding. This tool consists of a Data Flow Diagram CASE (DFD CASE) tool and a Structured chart PAD CASE tool to integrate a data flow diagram and a structured chart for each stage of design. Design of a hierarchical detailing is first expressed on algorithm using DFD, and then the control is designed and expressed by PAD. By combined operation between DFD CASE tool and PAD CASE tool, design knowledge of a hierarchical detailing is shared to both DFD design and PAD design. Design knowledge is automatically acquired from respective documents and stored in the knowledge base. The acquired design knowledge may be reused in following designs. By reusing it, a similar software system may be designed automatically. This tool features universality, an essentially zero start-up cost for automatic design, aud a substantial increase in software productivity after enough experiences have been accumulated.
Original Language: English
 


 
   Research Doc. 1999-10

Click to download PDF version

Authors(s):Mahmoud M. El-Khouly, Behrouz H. Far, Zenya Koono
Title:
"Software Agents for Reuse Software Components"
Transactions/Magazines/Conferences:
Proc. Seventh International Conference on Artificial Intelligence Applications, ICAIA' 99, pp. 299-306, Cairo, Egypt, February, 1999.
Abstract
Many important and useful applications for software agents require multiple agents on a network that communicate with each other. Such agents can retrieve software components for reuse from their repositories. However, if the keywords of the retrieving disaccord, the retrieval can be failure even if the eligible components exist in the software components repository. We propose an information retrieval technique to help in retrieving the software components that an agent wants to use from other agents' software components repositories. This technique uses semantic check to compare between the retrieved component with the agent requirement. The advantage of this technique is that the exact match is not necessary to find a similar software component.
Original Language: English
 


 
   Research Doc. 1999-11

Click to download PDF version

Authors(s): Behrouz H. Far
Title:
"Organizational Knowledge and Its Application in Multiagent System Development"
Transactions/Magazines/Conferences:
電子情報通信学会技術研究報告 資料 (Technical Reports of IEICE), AI99-44,, pp. 23-30, Tokyo, Japan, September, 1999.
Abstract
本研究では、複数のエージェントから構成されるネットワーク上の組織 (GAG:Generalized Agency、分散型情報処理環境) と標準化されたエージェント(AG:Generalized Agent) のモデルを紹介する。GAG は、組織知識を持っている。組織知識は、 個々のエージェントが所有するものではなく、エージェントの相互作用から 発生する知識である。本研究では、記号構造による組織知識のモデル化、 獲得、応用について述べる。 様々な応用例が考えられるが、電子商取引システムを実験対象とする。
Although there are many projects focusing on multiagent systems, there are only a few focusing on systematic design of large scale multiagent system. In this paper we will formalize the knowledge representation and sharing of agents, using symbol structures, define agencies as organizations (i.e., a coalition of agents), propose a formalism to represent organizational Intelligence, devise a basic configuration for generalized agents (AG), and derive certain patterns for generalized agencies (GAG) in order to use them effectively in a large scale multiagent system design. The private knowledge of an AG agent is represented by symbol structure and AG agents can share their knowledge using combination, specialization and generalization methods. Opposite to the other works, organizational knowledge, is defined as a property of at least a pair of AG agents. GAG is defined as an organization of AG agents. Each GAG describes a problem, such as a business process that happens repeatedly, and describes the participant agents as well as the process towards the solution to the problem. Furthermore, alternative configurations, decisions and trade-offs are defined. Electronic commerce is an example of such systems.
Original Language: English
 


 
   Research Doc. 1999-12

Click to download PDF version

Authors(s): Behrouz H. Far
Title:
"How to Learn Function of Devices"
Transactions/Magazines/Conferences:
Proc. 1999 IEEE Annual International Conference on Systems, Man and Cybernetics, SMC' 99, pp. V-860 - V-865, Tokyo, Japan, October, 1999.
Abstract
A goal of this research is developing methods for learning function of devices from a qualitative description of their structure and behavior using functional reasoning. Qualitative Function Formation (QFF) technique is introduced. In QFF, a function concept is defined as an interpretation of a persistence or an order in the sequence of qualitative states, using trace of qualitative state vector derived by simulation on a qualitative model of a device. This puts together the viewpoints on qualitative model, behavior and function, allowing systematic derivation of function from structure and behavior. Examples of application of QFF in categorization and learning function of components in engineering design are given. QFF has been successfully tested and used in engineering design, tool utilization and fault diagnosis.
Original Language: English
 


 
   Research Doc. 1999-13

Click to download PDF version

Authors(s):Mahmoud M. El-Khouly, Behrouz H. Far, Zenya Koono
Title:
"A New Multi-level Information Retrieval Technique for Reuse Software Components"
Transactions/Magazines/Conferences:
Proc. 1999 IEEE Annual International Conference on Systems, Man and Cybernetics, SMC' 99, pp. V-773 - V-777, Tokyo, Japan, October, 1999.
Abstract
In automatic programming area, the users still fail to find available software components that match their needs faster than developing them again. In this paper, a new information retrieval technique to help in retrieving the software components from repositories based on different levels of accuracy (i.e., exact match, match, and similar) had been presented. The software components had been classified into classes according to their functionality, and then they had been stored with what we call 'data dictionary' in a repository. The exact match is not required to retrieve a required software component from this repository using our retrieving technique. An example of how to apply this technique in automatic programming area to produce a program code list automatically had been given. The feature of learning the structure of new software component had been existed.
Original Language: English
 


 
   Research Doc. 1999-14

Click to download PDF version

Authors(s):Behrouz H. Far,
Title:
"Internet Based Syllabuses System (WebSYL) Using Personal Assistant Agents"
Transactions/Magazines/Conferences:
Advanced Research in Computers and Communications in Education, G. Cumming et al. (Eds.), pp. 103 - 110, IOS Press, 1999.
Abstract
The WebSYL system for the Electronic Syllabuses of the Faculty of Engineering, Saitama University is developed. The system contains entries for the lectures of all of the 6 engineering departments. The WebSYL system includes programs to automatically convert the database to hypertext format, programs for automatic generation of syllabuses entries by the instructors, form based and frame based search facilities and a bilingual search engine for keyword search. An important problem with systems with dynamic contents is the overhead on system administration to keep the system up to date. An important feature of the WebSYL system is minimizing the administration, by totally automating the contents management, using personal assistance agents and distributed objects technology. In this way instructors can update syllabuses entries from remote clients using ordinary browsers and the state of the edit is preserved for future updates. Automating this process has reduced the manual maintenance costs to zero.
Original Language: English
 


 
   Research Doc. 1999-15

Click to download PDF version

Authors(s):M. El-khouly, Behrouz H. Far, Z. Koono,
Title:
"A Multi-Agent Internet Based Tutoring System (I-ATCL) for Teaching Computer Programming Languages"
Transactions/Magazines/Conferences:
Advanced Research in Computers and Communications in Education, G. Cumming et al. (Eds.), pp. 153 - 156, IOS Press, 1999.
Abstract
This paper presents an agent-based tutoring system (I-ATCL) in presence of agent and internet technology. The I-ATCL system consists of two agents; a personal assistant agent for teachers (PAA-T) and a personal assistant agent for students (PAA-S). The PAA-S resides on the client side and communicates via IIOP with the PAA-T on the server side. This structure allows customization of the PAA-S to the needs of the student without putting extra burden on the server. Also this allows having many teacher agents attending to the needs of a single or multiple student agents. The PAA-T allows the teachers to cope with the knowledge base of a computer language under investigation, add or modify the commands' structure that will be taught. Also, this agent can generate a new tutoring dialog for a new computer programming language by consulting previous tutoring dialogs for another computer programming language. PAA-S contains student model and tutoring module. In student model, the system can accept free-format answer from the student, and checks it against the language structure.
Original Language: English
 


 
   Research Doc. 1999-16

Click to download PDF version

Authors(s):ベルーズ ホマユン ファー、原 正史、南雲 浩二、高橋 一成、近藤 邦雄、八嶋 弘幸
Title:
"大学教育におけるインターネット上の電子シラバス(授業計画)システム:WebSYL"
Transactions/Magazines/Conferences:
工学教育 第47巻 第5号, pp. 38 - 42, 1999.
Abstract
本論文では、埼玉大学工学部電子シラバス(授業計画)WebSYLシステムの設計と運営について論じる。本システムでは、インターネット上で各講義科目のシラバスを検索・印刷できる。技術の面では、通常はインターネット上で状態を持ったシステム設計は極めて複雑であるが、本システムでは、エージェントと分散オブジェクト技術を用いて作業途中の状態を記録できるように設計されている。その結果、教官は遠隔端末から担当科目のシラバスを編集できる。また、優れたユーザーインターフェイスと拡張性を持つ設計が本システムの特徴である。
Original Language: Japanese
 


 
   Research Doc. 1999-17

Click to download PDF version

Authors(s):Shadan Saniepour, Behrouz H. Far
Title:
"Mining Adaptation Rules from Cases in CBR Systems"
Transactions/Magazines/Conferences:
Discovery Science, A. Arikawa and F. Furukawa (edts.), Lecture Notes in Artificial Intelligence 1721, pp. 355 - 356, 1999.
Original Language: English
 


University of Calgary Last modified:

fare-mailacm.org