P
U
B
L
I
S
H
E
D
P
A
P
E
R
S
2
0
0
1
|
 |
 |
 |
| |
|
Research Doc. 2001-3
|
|
- Authors(s):
Hiroyuki Onjo,
Behrouz H. Far
- Title:
-
"A Unified View of Heterogeneous Agents' Interaction"
- Transactions/Magazines/Conferences:
- Transactions of the IEICE, Special Issue on Software Agents,
Vol. E84-D, No. 8, pp. 945-956, August 2001.
- Abstract
- In this paper agents' interactions are defined in terms of
cooperation, coordination and competition.
As for cooperation and coordination problems, we focus on knowledge
sharing of agents, define agencies as organizations of agents,
propose a method to extract organizational knowledge
for interacting agents. In case of competition, knowledge sharing
is impossible. Therefore, modeling and formalization of strategic
decision making and uncertainty management is required.
We present an incomplete game theoretical based decision
making method for competitive agents.
- Original Language: English
|
|
|
 |
 |
 |
|
|
|
| |
|
Research Doc. 2001-6
|
|
- Authors(s):
Romi Satria Wahono, Behrouz H. Far
- Title:
-
"Towards The Use of Intelligent Agents in Collaborative Object-Oriented Analysis and Design"
- Transactions/Magazines/Conferences:
- The 15th Annual Conference of Japanese Society for Artificial Intelligence, JSAI' 2001, pp. -, International Session, Japan, May, 2001.
- Abstract
- Software design often requires collaborative working between members of a software design project team. In many cases, the members are geographically distributed making the need for effective information and communication technologies acute. Implementing distributed artificial intelligence in intelligent agents is an alternative approach to achieve tasks on distributed computer systems. This paper examines the issues associated with the use of intelligent agents within the software analysis and design, especially we concern to solve the problems identified in object model creation process for object-oriented analysis and design. This is intended to serve as a useful decision support system for designers, and should allow faster, better, and more economic, collaborative object-oriented analysis and design.
- Original Language: English
|
|
|
|
|
|
|
|
|
| |
|
Research Doc. 2001-7
|
|
- Authors(s):
Zenya Koono, Hui Chen, H. Abolhassani, Behrouz H. Far
- Title:
-
"Design Knowledge and Software Engineering"
- Transactions/Magazines/Conferences:
- Journal of Natural Sciences,, Vol. 6, No. 1-2,
pp. 46-58, 2001.
- Abstract
- This paper points out various relationships between Design Knowledge and Software Engineering.
After introduction of human design, the relationships between industrial Software Engineering is discussed,
then further details of human design knowledge are revealed with discussions on humanistic aspects of design.
Key words : software design; design knowledge; automatic software design; learning effect.
- Original Language: English
|
|
|
|
|
|
|
|
|
| |
|
Research Doc. 2001-9
|
|
- Authors(s):
Hassan Hajji, Behrouz H. Far, Jingde Cheng
- Title:
- "Detection of Network Faults and Performance Problems"
- Transactions/Magazines/Conferences:
- Proceedings of the Internet Conference 2001, (IC' 2001), Osaka, Japan., November 2001.
(Best Paper Award)
- Abstract
- Network normal operation baselining for automatic detection of anomalies is addressed.
A model for network traffic is presented in which studied variables are modeled as a finite mixture model.
Based on stochastic approximation of the maximum likelihood function, we propose a baseline of network
normal operation as the asymptotic distribution of the difference between successive estimates of model parameters.
The baseline multivariate random variable is shown to be stationary, with mean zero under normal operation.
Performance problems are characterized by sudden jumps in the mean.
Detection is formulated as an online change point, problem, where the task is to process
residuals and raise alarms as soon as anomalies occur.
An analytical expression of false alarm rate allows us to choose the threshold, automatically.
Extensive experimental results on a real network showed that, the monitoring agent is able to
detect even slight changes in the characteristics of the network, and adapt to traffic patterns,
while maintaining a low alarm rate. Despite large fluctuations in network traffic, this work proves
that tailoring traffic modeling to specific gaols can be efficiently achieved.
- Original Language: English
|
|
|
|
|
|
|
acm.org