A human-inspired collective intelligence model for multi-agent based system

Saraswathy Shamini Gunasekaran, M. S. Ahmad, S. A. Mostafa

Research output: Contribution to journalArticle

Abstract

The collaborative and competitive nature of multi-agent systems (MAS) is visible through the simple social mode of communication that emerges between human-agent interactions or agent-to-agent interactions. A simple mode of communication involves the fundamental actions carried out by individual agents in achieving their desired goal. The sum of these achievements contribute to the overall group goal. Comparatively, the collective intelligence (CI) of a MAS simply means that these agents should work together to produce better solutions than those made possible when using the traditional approach. In designing MAS with CI properties, formalisation of a higher level deliberation process is essential. A high level deliberation process refers to the judgement comprehension of tasks, reasoning and problem solving and planning. In this paper, we propose our Collective Intelligence Model, CIM, which has the potential to control and coordinate a high-level deliberation process of a MAS. CIM is inspired by the emerging processes of controlled discussion, argumentation and negotiation between two or more intelligent human agents. These processes screen and validate the deliberation process through a cross-fertilisation approach. The emergent property of the cross-fertilised ideas results in an intelligent solution that solves optimisation-related tasks.

Original languageEnglish
Pages (from-to)39-54
Number of pages16
JournalPertanika Journal of Science and Technology
Volume25
Issue numberS10
Publication statusPublished - 01 Oct 2017

Fingerprint

Intelligence
animal communication
Multi agent systems
communication
Communication
planning
Negotiating
Fertilization
Planning

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Chemical Engineering(all)
  • Environmental Science(all)
  • Agricultural and Biological Sciences(all)

Cite this

@article{8b886d771fae47d4bdbbe8fa411c6ced,
title = "A human-inspired collective intelligence model for multi-agent based system",
abstract = "The collaborative and competitive nature of multi-agent systems (MAS) is visible through the simple social mode of communication that emerges between human-agent interactions or agent-to-agent interactions. A simple mode of communication involves the fundamental actions carried out by individual agents in achieving their desired goal. The sum of these achievements contribute to the overall group goal. Comparatively, the collective intelligence (CI) of a MAS simply means that these agents should work together to produce better solutions than those made possible when using the traditional approach. In designing MAS with CI properties, formalisation of a higher level deliberation process is essential. A high level deliberation process refers to the judgement comprehension of tasks, reasoning and problem solving and planning. In this paper, we propose our Collective Intelligence Model, CIM, which has the potential to control and coordinate a high-level deliberation process of a MAS. CIM is inspired by the emerging processes of controlled discussion, argumentation and negotiation between two or more intelligent human agents. These processes screen and validate the deliberation process through a cross-fertilisation approach. The emergent property of the cross-fertilised ideas results in an intelligent solution that solves optimisation-related tasks.",
author = "Gunasekaran, {Saraswathy Shamini} and Ahmad, {M. S.} and Mostafa, {S. A.}",
year = "2017",
month = "10",
day = "1",
language = "English",
volume = "25",
pages = "39--54",
journal = "Pertanika Journal of Science and Technology",
issn = "0128-7680",
publisher = "Universiti Putra Malaysia",
number = "S10",

}

A human-inspired collective intelligence model for multi-agent based system. / Gunasekaran, Saraswathy Shamini; Ahmad, M. S.; Mostafa, S. A.

In: Pertanika Journal of Science and Technology, Vol. 25, No. S10, 01.10.2017, p. 39-54.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A human-inspired collective intelligence model for multi-agent based system

AU - Gunasekaran, Saraswathy Shamini

AU - Ahmad, M. S.

AU - Mostafa, S. A.

PY - 2017/10/1

Y1 - 2017/10/1

N2 - The collaborative and competitive nature of multi-agent systems (MAS) is visible through the simple social mode of communication that emerges between human-agent interactions or agent-to-agent interactions. A simple mode of communication involves the fundamental actions carried out by individual agents in achieving their desired goal. The sum of these achievements contribute to the overall group goal. Comparatively, the collective intelligence (CI) of a MAS simply means that these agents should work together to produce better solutions than those made possible when using the traditional approach. In designing MAS with CI properties, formalisation of a higher level deliberation process is essential. A high level deliberation process refers to the judgement comprehension of tasks, reasoning and problem solving and planning. In this paper, we propose our Collective Intelligence Model, CIM, which has the potential to control and coordinate a high-level deliberation process of a MAS. CIM is inspired by the emerging processes of controlled discussion, argumentation and negotiation between two or more intelligent human agents. These processes screen and validate the deliberation process through a cross-fertilisation approach. The emergent property of the cross-fertilised ideas results in an intelligent solution that solves optimisation-related tasks.

AB - The collaborative and competitive nature of multi-agent systems (MAS) is visible through the simple social mode of communication that emerges between human-agent interactions or agent-to-agent interactions. A simple mode of communication involves the fundamental actions carried out by individual agents in achieving their desired goal. The sum of these achievements contribute to the overall group goal. Comparatively, the collective intelligence (CI) of a MAS simply means that these agents should work together to produce better solutions than those made possible when using the traditional approach. In designing MAS with CI properties, formalisation of a higher level deliberation process is essential. A high level deliberation process refers to the judgement comprehension of tasks, reasoning and problem solving and planning. In this paper, we propose our Collective Intelligence Model, CIM, which has the potential to control and coordinate a high-level deliberation process of a MAS. CIM is inspired by the emerging processes of controlled discussion, argumentation and negotiation between two or more intelligent human agents. These processes screen and validate the deliberation process through a cross-fertilisation approach. The emergent property of the cross-fertilised ideas results in an intelligent solution that solves optimisation-related tasks.

UR - http://www.scopus.com/inward/record.url?scp=85049143080&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85049143080&partnerID=8YFLogxK

M3 - Article

VL - 25

SP - 39

EP - 54

JO - Pertanika Journal of Science and Technology

JF - Pertanika Journal of Science and Technology

SN - 0128-7680

IS - S10

ER -