TeamSkill evolved

Mixed classification schemes for team-based multi-player games

Colin DeLong, Jaideep Srivastava

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

4 Citations (Scopus)

Abstract

In this paper, we introduce several approaches for maintaining weights over the aggregate skill ratings of subgroups of teams during the skill assessment process and extend our earlier work in this area to include game-specific performance measures as features alongside aggregate skill ratings as part of the online prediction task. We find that the inclusion of these game-specific measures do not improve prediction accuracy in the general case, but do when competing teams are considered evenly matched. As such, we develop a "mixed" classification method called TeamSkill-EVMixed which selects a classifier based on a threshold determined by the prior probability of one team defeating another. This mixed classification method outperforms all previous approaches in most evaluation settings and particularly so in tournament environments. We also find that TeamSkill-EVMixed's ability to perform well in close games is especially useful early on in the rating process where little game history is available.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages26-37
Number of pages12
Volume7301 LNAI
EditionPART 1
DOIs
Publication statusPublished - 2012
Externally publishedYes
Event16th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining, PAKDD 2012 - Kuala Lumpur
Duration: 29 May 20121 Jun 2012

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 1
Volume7301 LNAI
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other16th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining, PAKDD 2012
CityKuala Lumpur
Period29/5/121/6/12

Fingerprint

Game
Classifiers
Prior Probability
Prediction
Tournament
Performance Measures
Inclusion
Classifier
Subgroup
Evaluation
Skills
History

Keywords

  • competitive gaming
  • confidence-weighted learning
  • passive aggressive algorithm
  • perceptron
  • Player rating systems

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

DeLong, C., & Srivastava, J. (2012). TeamSkill evolved: Mixed classification schemes for team-based multi-player games. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (PART 1 ed., Vol. 7301 LNAI, pp. 26-37). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7301 LNAI, No. PART 1). https://doi.org/10.1007/978-3-642-30217-6_3

TeamSkill evolved : Mixed classification schemes for team-based multi-player games. / DeLong, Colin; Srivastava, Jaideep.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7301 LNAI PART 1. ed. 2012. p. 26-37 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7301 LNAI, No. PART 1).

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

DeLong, C & Srivastava, J 2012, TeamSkill evolved: Mixed classification schemes for team-based multi-player games. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 1 edn, vol. 7301 LNAI, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 1, vol. 7301 LNAI, pp. 26-37, 16th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining, PAKDD 2012, Kuala Lumpur, 29/5/12. https://doi.org/10.1007/978-3-642-30217-6_3
DeLong C, Srivastava J. TeamSkill evolved: Mixed classification schemes for team-based multi-player games. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 1 ed. Vol. 7301 LNAI. 2012. p. 26-37. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1). https://doi.org/10.1007/978-3-642-30217-6_3
DeLong, Colin ; Srivastava, Jaideep. / TeamSkill evolved : Mixed classification schemes for team-based multi-player games. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7301 LNAI PART 1. ed. 2012. pp. 26-37 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1).
@inproceedings{e0d489432026450aa95a6144b3fe3400,
title = "TeamSkill evolved: Mixed classification schemes for team-based multi-player games",
abstract = "In this paper, we introduce several approaches for maintaining weights over the aggregate skill ratings of subgroups of teams during the skill assessment process and extend our earlier work in this area to include game-specific performance measures as features alongside aggregate skill ratings as part of the online prediction task. We find that the inclusion of these game-specific measures do not improve prediction accuracy in the general case, but do when competing teams are considered evenly matched. As such, we develop a {"}mixed{"} classification method called TeamSkill-EVMixed which selects a classifier based on a threshold determined by the prior probability of one team defeating another. This mixed classification method outperforms all previous approaches in most evaluation settings and particularly so in tournament environments. We also find that TeamSkill-EVMixed's ability to perform well in close games is especially useful early on in the rating process where little game history is available.",
keywords = "competitive gaming, confidence-weighted learning, passive aggressive algorithm, perceptron, Player rating systems",
author = "Colin DeLong and Jaideep Srivastava",
year = "2012",
doi = "10.1007/978-3-642-30217-6_3",
language = "English",
isbn = "9783642302169",
volume = "7301 LNAI",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
number = "PART 1",
pages = "26--37",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
edition = "PART 1",

}

TY - GEN

T1 - TeamSkill evolved

T2 - Mixed classification schemes for team-based multi-player games

AU - DeLong, Colin

AU - Srivastava, Jaideep

PY - 2012

Y1 - 2012

N2 - In this paper, we introduce several approaches for maintaining weights over the aggregate skill ratings of subgroups of teams during the skill assessment process and extend our earlier work in this area to include game-specific performance measures as features alongside aggregate skill ratings as part of the online prediction task. We find that the inclusion of these game-specific measures do not improve prediction accuracy in the general case, but do when competing teams are considered evenly matched. As such, we develop a "mixed" classification method called TeamSkill-EVMixed which selects a classifier based on a threshold determined by the prior probability of one team defeating another. This mixed classification method outperforms all previous approaches in most evaluation settings and particularly so in tournament environments. We also find that TeamSkill-EVMixed's ability to perform well in close games is especially useful early on in the rating process where little game history is available.

AB - In this paper, we introduce several approaches for maintaining weights over the aggregate skill ratings of subgroups of teams during the skill assessment process and extend our earlier work in this area to include game-specific performance measures as features alongside aggregate skill ratings as part of the online prediction task. We find that the inclusion of these game-specific measures do not improve prediction accuracy in the general case, but do when competing teams are considered evenly matched. As such, we develop a "mixed" classification method called TeamSkill-EVMixed which selects a classifier based on a threshold determined by the prior probability of one team defeating another. This mixed classification method outperforms all previous approaches in most evaluation settings and particularly so in tournament environments. We also find that TeamSkill-EVMixed's ability to perform well in close games is especially useful early on in the rating process where little game history is available.

KW - competitive gaming

KW - confidence-weighted learning

KW - passive aggressive algorithm

KW - perceptron

KW - Player rating systems

UR - http://www.scopus.com/inward/record.url?scp=84861446458&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84861446458&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-30217-6_3

DO - 10.1007/978-3-642-30217-6_3

M3 - Conference contribution

SN - 9783642302169

VL - 7301 LNAI

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 26

EP - 37

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -