[논문]기계학습을 활용한 게임승패 예측 및 변수중요도 산출을 통한 전략방향 도출

김용우; 김영민

doi:10.7583/jkgs.2021.21.4.3

[국내논문] 기계학습을 활용한 게임승패 예측 및 변수중요도 산출을 통한 전략방향 도출
Predicting Game Results using Machine Learning and Deriving Strategic Direction from Variable Importance 원문보기

한국게임학회 논문지 = Journal of Korea Game Society, v.21 no.4, 2021년, pp.3 - 12

김용우 (한양대학교 기술경영대학원) , 김영민 (한양대학교 기술경영대학원)

초록
AI-Helper

본 연구에서는 게임 초반 10분의 데이터를 이용하여 리그오브레전드 게임의 최종승패를 랭크별로 예측하고, 구축된 승패예측 모형으로부터 변수중요도를 추출하여 승리를 위한 초반 게임운영의 방향성을 알아보았다. 그 결과 모든 랭크에서 70% 이상의 정확도로 승패를 예측할 수 있었다. 이는 경기 양상이 대부분 뒤집히지 않고 최종승패로 이어지는 것을 의미하며, 이러한 경향성은 상위 랭크로 갈수록 더욱 강하게 나타났다. 랭크와 무관하게 킬(데스)가 초반 게임에서 최종승패에 가장 큰 영향을 미치는 요소로 나타났으나, 일부 변수는 랭크에 따라 중요도 순위가 변화하였고 이는 유저가 속한 랭크에 따라 승리에 효과적인 초반 전략방향에 차이가 있음을 시사한다.

Abstract ▼ AI-Helper

In this study, models for predicting the final result of League of Legends game were constructed for each rank using data from the first 10 minutes of the game. Variable importance was extracted from the prediction models to derive strategic direction in early phase of the game. As a result, it was possible to predict final results with over 70% accuracy in all ranks. It was found that early game advantage tends to lead to the final win and this tendency appeared stronger as it goes to challenger ranks. Kill(death) was found to be the most influential factor for win, however, there were also variables whose importance rank changed according to rank. This indicates there is a difference in the strategic direction in the early stage of the game depending on the rank.

주제어

표/그림 (5)

표 [Table 1] List of Variables
그림 [Fig. 1] Process of predicting game results and extracting variable importance by rank
표 [Table 2] Model Performance
표 [Table 3] TOP5 important variables
표 [Table 4] Variable Importance

참고문헌 (26)

Korea Creative Content Agency, "The 2020 Survey on the Korean e-Sports Industry", Korea Creative Content Agency, 2020.
Sae-Sook Oh and Dae-Hoon Kim, "Analysis of the Academic Research Trend of e-sports", Journal of Wellness, vol. 7, No. 2, pp.113-121, 2012.
Jung Hwan Cho, "Utilization and Prospect of Sport Big Data", Korean Society For Measurement And Evaluation In Physical Education And Sports Science, vol. 14, No. 3, pp.01-11, 2012.

상세보기
Yong Goo Kang and Huy Kang Kim, "Game Bot Detection Based on Action Time Interval", Journal of The Korea Institute of Information Security and Cryptology, vol. 28, No. 5, pp.1153-1160, 2018.

원문보기 상세보기
GGtics, "Analytics/AI to help players win - YOUR.GG", accessed Apr 13, 2021, https://your.gg/?langen.
OP.GG, "LoL Stats, Record Replay, Database, Guide - OP.GG", accessed Apr 13, 2021, https://www.op.gg.
일요서울, "사교육 시장에 등장한 '게임학원', 그 배경은?", accessed Apr 13, 2021, http://www.ilyoseoul.co.kr/news/articleView.html?idxno237470.
Alexander Neumann, "Developing a Model to Predict Match Outcomes in League of Legends," Barrett, The Honors College Thesis, Arizona State University, 2015.
Cheolgi Kim and Soowon Lee, "Predicting Win-Loss of League of Legends Using Bidirectional LSTM Embedding", Korea Information Processing Society, vol. 9, No. 2, pp.61-68, 2020.
Dong-Wook Kim, Jeawon Park and Jaehyun Choi, "A study for the prediction of winning of e-Sports using machine Learning", Jounal of The Korea Society of Information Technology Policy & Management, vol. 9, No. 1, pp.319-325, 2017.
Jimin Ku and Jaehee Kim, "Development of game indicators and winning forecasting models with game data", Journal of the Korean Data and Information Science Society, vol. 28, no. 2, pp.237-250, 2017.

원문보기 상세보기
Min-ji Oh, Eun-seon Choi, Som Akhamixay Oui and Wan-sup Cho, "Predicting win-loss using game data and deriving the importance of subdivided variables", The Korea Journal of BigData, vol. 5, No. 2, pp.231-240, 2020.

원문보기 상세보기
Ani, R., Harikumar, V., Devan, A. K., and Deepa, O. S., "Victory prediction in League of Legends using Feature Selection and Ensemble methods", In 2019 International Conference on Intelligent Computing and Control Systems, pp. 74-77, 2018.
Thompson JJ, Blair MR, Chen L and Henrey AJ, "Video Game Telemetry as a Critical Tool in the Study of Complex Skill Learning", PLoS ONE 8(9): e75129.

상세보기
Lee, Sang-Kwang, Seung-Jin Hong, and Seong-Il Yang. "Predicting Game Outcome in Multiplayer Online Battle Arena Games", International Conference on Information and Communication Technology Convergence, IEEE, 2019.
Riot Games, "Riot Developer Portal", accesse d Apr 13, 2021, https://developer.riotgames.com.
Gregorutti B., Michel B. and Saint-Pierre P, "Correlation and variable importance in random forests", Statistics and Computing, vol. 27, pp.659-678, 2017.

상세보기
Bzdok, D., Altman, N., and Krzywinski, M., "Statistics versus machine learning", Nature methods, 15(4), pp 233-234, 2018.

상세보기
Rich Caruana and Alexandru Niculescu-Mizil, "An empirical comparison of supervised learning algorithms", In Proceedings of the 23rd international conference on Machine learning, pp.161-168, 2006.
Bentejac C., Csorgo, A. and Martinez-Munoz G., "A comparative analysis of gradient boosting algorithms", Artificial Intelligence Review, vol. 54, pp.1937-1967, 2021.

상세보기
Van Saeys, Inaki Inza and Pedro Larranaga, "A review of feature selection techniques in bioinformatics", Bioinformatics, Volume 23, Issue 19, pp. 2507-2517, 2007.

상세보기
Nicodemus K.K., "Letter to the Editor: On the stability and ranking of predictors from random forest variable importance measures", Brief Bioinform, vol. 12, No. 4, pp.369-373, 2011.

상세보기
Silva, A., G. Pappa and L. Chaimowicz, "Continuous Outcome Prediction of League of Legends Competitive Matches Using Recurrent Neural Networks.", In Proceedings of SBGames, code: 188226, 2018.
Ellis Cashmore. "Making Sense of Sports", Routledge, 2010.
Seong-Eun Seo and Chi-Yo Kim, "Recognition of the Type and Cause of Trolling", Korea Game Society, vol. 15, No. 4, pp. 93-110, 2015.
Kyu Bok Lee and Young Jae Kim, "Analysis of Factors that Influence Users' Preference for MOBA Game Genre: Focusing on the Game Systems of League of Legends", Global Cultural Contents, vol. 47, pp. 107-124, 2021.

내보내기 구분	파일저장 인쇄 메일전송
구성항목	기본정보 상세정보 관리번호, 논문명, 저널/프로시딩명, 저자 , 발행년, 권, 호, 시작페이지, 끝페이지, 발행기관 관리번호, 논문명, 대등논문명, 저자 , 저널/프로시딩명, 발행기관, 발행년, 발행언어, 권, 호, 시작페이지, 끝페이지, ISBN, ISSN, 주제분야, 키워드, 초록(한글), 초록(영문), 저자(소속기관)
저장형식	Text(ASCII format) Excel format RefWorks Direct Export RIS format (for Reference Manager, ProCite, EndNote), Scholar's Aids, Mendeley
메일정보	받는사람 (필수) @ 보내는사람 (선택) @ 제목 내용 KISTI 검색결과 이메일 서비스
안내	총 건의 자료가 검색되었습니다. 다운받으실 자료의 인덱스를 입력하세요. (1-10,000) 검색결과의 순서대로 최대 10,000건 까지 다운로드가 가능합니다. 데이타가 많을 경우 속도가 느려질 수 있습니다.(최대 2~3분 소요) 다운로드 파일은 UTF-8 형태로 저장됩니다. 파일의 내용이 제대로 보이지 않을실 때는 웹브라우저 상단의 보기 -> 인코딩 -> 자동선택 여부를 확인하십시오. ~ Text(ASCII format) Excel format

연합인증