Exploiting Domain Knowledge in Making Delegation Decisions

Chukwuemeka David Emele, Timothy J Norman, Murat Sensoy, Simon Parsons

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)
4 Downloads (Pure)

Abstract

In multi-agent systems, agents often depend on others to act on their behalf. However, delegation decisions are complicated in norm- governed environments, where agents’ activities are regulated by policies. Especially when such policies are not public, learning these policies be- come critical to estimate the outcome of delegation decisions. In this paper, we propose the use of domain knowledge in aiding the learning of policies. Our approach combines ontological reasoning, machine learning and argumentation in a novel way for identifying, learning, and modeling policies. Using our approach, software agents can autonomously reason about the policies that others are operating with, and make informed decisions about to whom to delegate a task. In a set of experiments, we demonstrate the utility of this novel combination of techniques through empirical evaluation. Our evaluation shows that more accurate models of others’ policies can be developed more rapidly using various forms of domain knowledge.
Original languageEnglish
Title of host publicationAgents and Data Mining Interaction
Subtitle of host publication7th International Workshop on Agents and Data Mining Interation, ADMI 2011, Taipei, Taiwan, May 2-6, 2011, Revised Selected Papers
EditorsLongbing Cao, Ana L.C. Bazzan, Andreas L. Symeonidis, Vladimir I. Gorodetsky, Gerhard Weiss, Philip S. Yu
PublisherSpringer
Pages117-131
Number of pages15
Volume7103
ISBN (Print)978-3-642-27608-8
DOIs
Publication statusPublished - 19 Dec 2011
Event7th International Workshop on Agents and Data Mining Interation, ADMI 2011 - Taipeii, Taiwan, Province of China
Duration: 2 May 20116 May 2011

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
Volume7103
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference7th International Workshop on Agents and Data Mining Interation, ADMI 2011
CountryTaiwan, Province of China
CityTaipeii
Period2/05/116/05/11

Fingerprint

Software agents
Multi agent systems
Learning systems
Decision making
Experiments

Cite this

Emele, C. D., Norman, T. J., Sensoy, M., & Parsons, S. (2011). Exploiting Domain Knowledge in Making Delegation Decisions. In L. Cao, A. L. C. Bazzan, A. L. Symeonidis, V. I. Gorodetsky, G. Weiss, & P. S. Yu (Eds.), Agents and Data Mining Interaction: 7th International Workshop on Agents and Data Mining Interation, ADMI 2011, Taipei, Taiwan, May 2-6, 2011, Revised Selected Papers (Vol. 7103, pp. 117-131). (Lecture Notes in Computer Science; Vol. 7103). Springer . https://doi.org/10.1007/978-3-642-27609-5_9

Exploiting Domain Knowledge in Making Delegation Decisions. / Emele, Chukwuemeka David; Norman, Timothy J; Sensoy, Murat; Parsons, Simon.

Agents and Data Mining Interaction: 7th International Workshop on Agents and Data Mining Interation, ADMI 2011, Taipei, Taiwan, May 2-6, 2011, Revised Selected Papers. ed. / Longbing Cao; Ana L.C. Bazzan; Andreas L. Symeonidis; Vladimir I. Gorodetsky; Gerhard Weiss; Philip S. Yu. Vol. 7103 Springer , 2011. p. 117-131 (Lecture Notes in Computer Science; Vol. 7103).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Emele, CD, Norman, TJ, Sensoy, M & Parsons, S 2011, Exploiting Domain Knowledge in Making Delegation Decisions. in L Cao, ALC Bazzan, AL Symeonidis, VI Gorodetsky, G Weiss & PS Yu (eds), Agents and Data Mining Interaction: 7th International Workshop on Agents and Data Mining Interation, ADMI 2011, Taipei, Taiwan, May 2-6, 2011, Revised Selected Papers. vol. 7103, Lecture Notes in Computer Science, vol. 7103, Springer , pp. 117-131, 7th International Workshop on Agents and Data Mining Interation, ADMI 2011, Taipeii, Taiwan, Province of China, 2/05/11. https://doi.org/10.1007/978-3-642-27609-5_9
Emele CD, Norman TJ, Sensoy M, Parsons S. Exploiting Domain Knowledge in Making Delegation Decisions. In Cao L, Bazzan ALC, Symeonidis AL, Gorodetsky VI, Weiss G, Yu PS, editors, Agents and Data Mining Interaction: 7th International Workshop on Agents and Data Mining Interation, ADMI 2011, Taipei, Taiwan, May 2-6, 2011, Revised Selected Papers. Vol. 7103. Springer . 2011. p. 117-131. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-642-27609-5_9
Emele, Chukwuemeka David ; Norman, Timothy J ; Sensoy, Murat ; Parsons, Simon. / Exploiting Domain Knowledge in Making Delegation Decisions. Agents and Data Mining Interaction: 7th International Workshop on Agents and Data Mining Interation, ADMI 2011, Taipei, Taiwan, May 2-6, 2011, Revised Selected Papers. editor / Longbing Cao ; Ana L.C. Bazzan ; Andreas L. Symeonidis ; Vladimir I. Gorodetsky ; Gerhard Weiss ; Philip S. Yu. Vol. 7103 Springer , 2011. pp. 117-131 (Lecture Notes in Computer Science).
@inproceedings{d1398a1338da4b5da1426259ca94921e,
title = "Exploiting Domain Knowledge in Making Delegation Decisions",
abstract = "In multi-agent systems, agents often depend on others to act on their behalf. However, delegation decisions are complicated in norm- governed environments, where agents’ activities are regulated by policies. Especially when such policies are not public, learning these policies be- come critical to estimate the outcome of delegation decisions. In this paper, we propose the use of domain knowledge in aiding the learning of policies. Our approach combines ontological reasoning, machine learning and argumentation in a novel way for identifying, learning, and modeling policies. Using our approach, software agents can autonomously reason about the policies that others are operating with, and make informed decisions about to whom to delegate a task. In a set of experiments, we demonstrate the utility of this novel combination of techniques through empirical evaluation. Our evaluation shows that more accurate models of others’ policies can be developed more rapidly using various forms of domain knowledge.",
author = "Emele, {Chukwuemeka David} and Norman, {Timothy J} and Murat Sensoy and Simon Parsons",
note = "@inproceedings{conf/admi/EmeleNSP11, added-at = {2011-12-19T00:00:00.000+0100}, author = {Emele, Chukwuemeka David and Norman, Timothy J. and Sensoy, Murat and Parsons, Simon}, biburl = {http://www.bibsonomy.org/bibtex/20a08b683088443f1fd36d6ef28bf6615/dblp}, booktitle = {ADMI}, crossref = {conf/admi/2011}, editor = {Cao, Longbing and Bazzan, Ana L. C. and Symeonidis, Andreas L. and Gorodetsky, Vladimir and Weiss, Gerhard and Yu, Philip S.}, ee = {http://dx.doi.org/10.1007/978-3-642-27609-5_9}, interhash = {1d7e7f8554e8bdb3d43c32e02aeabcec}, intrahash = {0a08b683088443f1fd36d6ef28bf6615}, isbn = {978-3-642-27608-8}, keywords = {dblp}, pages = {117-131}, publisher = {Springer}, series = {Lecture Notes in Computer Science}, timestamp = {2011-12-19T00:00:00.000+0100}, title = {Exploiting Domain Knowledge in Making Delegation Decisions.}, url = {http://dblp.uni-trier.de/db/conf/admi/admi2011.html#EmeleNSP11}, volume = 7103, year = 2011 }",
year = "2011",
month = "12",
day = "19",
doi = "10.1007/978-3-642-27609-5_9",
language = "English",
isbn = "978-3-642-27608-8",
volume = "7103",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "117--131",
editor = "Longbing Cao and Bazzan, {Ana L.C.} and Symeonidis, {Andreas L. } and Gorodetsky, {Vladimir I. } and Gerhard Weiss and Yu, {Philip S. }",
booktitle = "Agents and Data Mining Interaction",

}

TY - GEN

T1 - Exploiting Domain Knowledge in Making Delegation Decisions

AU - Emele, Chukwuemeka David

AU - Norman, Timothy J

AU - Sensoy, Murat

AU - Parsons, Simon

N1 - @inproceedings{conf/admi/EmeleNSP11, added-at = {2011-12-19T00:00:00.000+0100}, author = {Emele, Chukwuemeka David and Norman, Timothy J. and Sensoy, Murat and Parsons, Simon}, biburl = {http://www.bibsonomy.org/bibtex/20a08b683088443f1fd36d6ef28bf6615/dblp}, booktitle = {ADMI}, crossref = {conf/admi/2011}, editor = {Cao, Longbing and Bazzan, Ana L. C. and Symeonidis, Andreas L. and Gorodetsky, Vladimir and Weiss, Gerhard and Yu, Philip S.}, ee = {http://dx.doi.org/10.1007/978-3-642-27609-5_9}, interhash = {1d7e7f8554e8bdb3d43c32e02aeabcec}, intrahash = {0a08b683088443f1fd36d6ef28bf6615}, isbn = {978-3-642-27608-8}, keywords = {dblp}, pages = {117-131}, publisher = {Springer}, series = {Lecture Notes in Computer Science}, timestamp = {2011-12-19T00:00:00.000+0100}, title = {Exploiting Domain Knowledge in Making Delegation Decisions.}, url = {http://dblp.uni-trier.de/db/conf/admi/admi2011.html#EmeleNSP11}, volume = 7103, year = 2011 }

PY - 2011/12/19

Y1 - 2011/12/19

N2 - In multi-agent systems, agents often depend on others to act on their behalf. However, delegation decisions are complicated in norm- governed environments, where agents’ activities are regulated by policies. Especially when such policies are not public, learning these policies be- come critical to estimate the outcome of delegation decisions. In this paper, we propose the use of domain knowledge in aiding the learning of policies. Our approach combines ontological reasoning, machine learning and argumentation in a novel way for identifying, learning, and modeling policies. Using our approach, software agents can autonomously reason about the policies that others are operating with, and make informed decisions about to whom to delegate a task. In a set of experiments, we demonstrate the utility of this novel combination of techniques through empirical evaluation. Our evaluation shows that more accurate models of others’ policies can be developed more rapidly using various forms of domain knowledge.

AB - In multi-agent systems, agents often depend on others to act on their behalf. However, delegation decisions are complicated in norm- governed environments, where agents’ activities are regulated by policies. Especially when such policies are not public, learning these policies be- come critical to estimate the outcome of delegation decisions. In this paper, we propose the use of domain knowledge in aiding the learning of policies. Our approach combines ontological reasoning, machine learning and argumentation in a novel way for identifying, learning, and modeling policies. Using our approach, software agents can autonomously reason about the policies that others are operating with, and make informed decisions about to whom to delegate a task. In a set of experiments, we demonstrate the utility of this novel combination of techniques through empirical evaluation. Our evaluation shows that more accurate models of others’ policies can be developed more rapidly using various forms of domain knowledge.

U2 - 10.1007/978-3-642-27609-5_9

DO - 10.1007/978-3-642-27609-5_9

M3 - Conference contribution

SN - 978-3-642-27608-8

VL - 7103

T3 - Lecture Notes in Computer Science

SP - 117

EP - 131

BT - Agents and Data Mining Interaction

A2 - Cao, Longbing

A2 - Bazzan, Ana L.C.

A2 - Symeonidis, Andreas L.

A2 - Gorodetsky, Vladimir I.

A2 - Weiss, Gerhard

A2 - Yu, Philip S.

PB - Springer

ER -