A Unified View of Software Agents Interactions

Behrouz Homayoun FAR; Wei WU; Mohsen AFSHARCHI

A Unified View of Software Agents Interactions

Behrouz Homayoun FAR, Wei WU, Mohsen AFSHARCHI

Full Text Views

0

Share
Cite this

Summary :

Software agents are knowledgeable, autonomous, situated and interactive software entities. Agents' interactions are of special importance when a group of agents interact with each other to solve a problem that is beyond the capability and knowledge of each individual. Efficiency, performance and overall quality of the multi-agent applications depend mainly on how the agents interact with each other effectively. In this paper, we suggest an agent model by which we can clearly distinguish different agent's interaction scenarios. The model has five attributes: goal, control, interface, identity and knowledge base. Using the model, we analyze and describe possible scenarios; devise the appropriate reasoning and decision making techniques for each scenario; and build a library of reasoning and decision making modules that can be used readily in the design and implementation of multiagent systems.

Publication: IEICE TRANSACTIONS on Information Vol.E87-D No.4 pp.896-907

Publication Date: 2004/04/01

Publicized

Online ISSN

DOI

Type of Manuscript: Special Section PAPER (Special Section on Knowledge-Based Software Engineering)

Category: Knowledge Engineering and Robotics

Cite this

Copy

Behrouz Homayoun FAR, Wei WU, Mohsen AFSHARCHI, "A Unified View of Software Agents Interactions" in IEICE TRANSACTIONS on Information, vol. E87-D, no. 4, pp. 896-907, April 2004, doi: .
Abstract: Software agents are knowledgeable, autonomous, situated and interactive software entities. Agents' interactions are of special importance when a group of agents interact with each other to solve a problem that is beyond the capability and knowledge of each individual. Efficiency, performance and overall quality of the multi-agent applications depend mainly on how the agents interact with each other effectively. In this paper, we suggest an agent model by which we can clearly distinguish different agent's interaction scenarios. The model has five attributes: goal, control, interface, identity and knowledge base. Using the model, we analyze and describe possible scenarios; devise the appropriate reasoning and decision making techniques for each scenario; and build a library of reasoning and decision making modules that can be used readily in the design and implementation of multiagent systems.
URL: https://globals.ieice.org/en_transactions/information/10.1587/e87-d_4_896/_p

Copy

@ARTICLE{e87-d_4_896,
author={Behrouz Homayoun FAR, Wei WU, Mohsen AFSHARCHI, },
journal={IEICE TRANSACTIONS on Information},
title={A Unified View of Software Agents Interactions},
year={2004},
volume={E87-D},
number={4},
pages={896-907},
abstract={Software agents are knowledgeable, autonomous, situated and interactive software entities. Agents' interactions are of special importance when a group of agents interact with each other to solve a problem that is beyond the capability and knowledge of each individual. Efficiency, performance and overall quality of the multi-agent applications depend mainly on how the agents interact with each other effectively. In this paper, we suggest an agent model by which we can clearly distinguish different agent's interaction scenarios. The model has five attributes: goal, control, interface, identity and knowledge base. Using the model, we analyze and describe possible scenarios; devise the appropriate reasoning and decision making techniques for each scenario; and build a library of reasoning and decision making modules that can be used readily in the design and implementation of multiagent systems.},
keywords={},
doi={},
ISSN={},
month={April},}

Copy

TY - JOUR
TI - A Unified View of Software Agents Interactions
T2 - IEICE TRANSACTIONS on Information
SP - 896
EP - 907
AU - Behrouz Homayoun FAR
AU - Wei WU
AU - Mohsen AFSHARCHI
PY - 2004
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E87-D
IS - 4
JA - IEICE TRANSACTIONS on Information
Y1 - April 2004
AB - Software agents are knowledgeable, autonomous, situated and interactive software entities. Agents' interactions are of special importance when a group of agents interact with each other to solve a problem that is beyond the capability and knowledge of each individual. Efficiency, performance and overall quality of the multi-agent applications depend mainly on how the agents interact with each other effectively. In this paper, we suggest an agent model by which we can clearly distinguish different agent's interaction scenarios. The model has five attributes: goal, control, interface, identity and knowledge base. Using the model, we analyze and describe possible scenarios; devise the appropriate reasoning and decision making techniques for each scenario; and build a library of reasoning and decision making modules that can be used readily in the design and implementation of multiagent systems.
ER -