KNN(K-Nearest Neighbors) 분류 알고리즘을 이용한 고교야구 선수의 프로야구 지명 예측기법
Professional Baseball Nomination Prediction Method for High School Baseball Players Using KNN(K-Nearest Neighbors) Classification Algorithm
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
4차 산업혁명은 사물인터넷(IoT)과 빅데이터(Big data)의 발전으로 가속화되고 있으며, 이를 이끄는 핵심 기술로서 인공지능(AI)과 기계 학습(ML)이 주목받고 있다. AI는 의료, 교육, 제조, 금융, 서비스 등 다양한 산업 분야에서 인공지능 기술이 핵심적인 역할을 수행할 것이고 새로운 서비스 개발, 생산성 향상, 비용 절감, 인적 오류 감소 등을 통해 사회 전반에 긍정적인 영향을 미칠 것으로 기대된다. 기계 학습은 방대한 양의 데이터를 분석하고 패턴을 파악함으로써 인공지능 시스템의 학습 능력을 강화하고, 더욱 정확하고 효과적인 의사 결정을 수행하도록 도울 수 있다. 야구 경기에서 좋은 투수의 기준은 다음과 같다. 기본적인 평가 지표로서 승리, 방어율, 탈삼진율, 한 이닝당 허용하는 피안타와 볼넷 개수의 합(WHIP)이 있다. 이러한 수치 외에도 뛰어난 제구력, 강력한 속구, 다양한 변화구, 강한 멘탈리티 그리고 리더십 등도 좋은 투수를 평가하는 기준이 될 수 있다. 본 논문에서는 Google colab을 이용하여 고등학교 투수의 프로야구 지명을 예측하기 위한 KNN 지도학습 분류 모델 기법을 연구한다. 이를 위하여 투수의 방어율과 WHIP를 학습데이터로 이용하며 예측 기법을 구현하고 성능을 평가한다. 이를 토대로 하여 어떤 투수가 프로야구에 지명되는 지를 예측할 수 있다.
Abstract
The 4th Industrial Revolution is accelerating with the development of the Internet of Things(IoT) and Big Data, and artificial intelligence(AI) and machine learning(ML) are attracting attention as core technologies. AI technology will play a key role in various industries such as healthcare, education, manufacturing, finance, and services, and will have a positive impact on society as a whole through the development of new services, improved productivity, cost reduction, and reduction of human errors. Machine learning can strengthen the learning ability of artificial intelligence systems and help them make more accurate and effective decisions by analyzing amounts of data and identifying patterns. The criteria for a good pitcher in a baseball game are as follows: Basic evaluation indicators include wins, ERA(Earned run average), strikeout rate, and the sum of hits and walks allowed per inning(WHIP). In addition to these numbers, excellent ball control, powerful fastballs, various breaking balls, strong mentality, and leadership can also be criteria for evaluating a good pitcher. In this paper, we study the KNN(K-Nearest Neighbors) supervised learning classification model method to predict professional baseball nominations of high school pitchers using Google colab. To achieve this, we use the pitcher's ERA and WHIP as training data, implement prediction techniques, and evaluate performance. Based on this, it is possible to predict which pitcher will be nominated to professional baseball.
참고문헌 (22)
- "Internet of Thing(IoT), Big Data, Artificial Intelligence(AI)" (2019)[URL]
- Introduction to machine learning, "J. Nilsson", Robotics Laboratory, Department of Computer Science, Stanford University, ~ (1998)
- C. Bishop, "Pattern recognition and machine learning" (2006), Springer
- T. Mitchell, "Machine learning" (1997), McGraw Hill
- "Supervised learning" (2023)[URL]
- I. Goodfellow, "Deep learning" (2016), MIT Press
- Y. H. Oh, "High-school baseball pitcher’s pitching speed prediction using linear regression analysis method", Journal of Knowledge Information Technology and Systems, 14(4), pp.381~390 (2019)
- Y. H. Oh, "High-school baseball pitcher’s ERA(earned run average)prediction using multi-variable linear regression analysis method", Journal of Knowledge Information Technology and Systems, 15(4), pp.497~506 (2020)
- Y. H. Oh, "A study on pro-baseball player's pomination prediction of high-school baseball pitcher using logistic regression classification method", Journal of Knowledge Information Technology and Systems, 16(4), pp.721~730 (2021)
- V. G. Costa, "Recent advances in decision trees : an updated survey", Artificial Intelligence Review, 56, pp.4765~4800 (2023)
- N. Ali, "Evaluation of k-nearest neighbour classifier performance for heterogeneous data sets", SN Applied Science, 1, pp.1559~ (2019)
- Y-J. Lee, "The kinematic analysis of the pitching motion for the straight and curve ball", Journal of The Korean Society of Sports Biomechanics, 12(2), pp.109~130 (2002)
- "Baseball Reference" (2022)[URL]
- "Fangraph Baseball" (2023)[URL]
- "Korea Baseball Softball Association" (2024)[URL]
- "Korea University Softball Foundation" (2024)[URL]
- J-Y. Baek, "A case study on the successful port marketing strategies of the korean professional baseball teams", Department of Convergence Industry, pp.2016
- Y. He, "Revisiting K-nearest neighbors for classification" (2020), Artificial Intelligence
- D. Wang, "K-nearest neighbors : A survey and extension", Proceeding of IEEE Transactions on Pattern Analysis and Machine Intelligence, 43(6), pp.1773~1790 (2021)
- D. Chang, "The devil is in the channels: Mutual-channel loss for fine-grained image classification", Proceeding of IEEE Transactions on Image Processing, pp.4683~4695 (2020)
- Y. H. Oh, "A study on pro-baseball player nomination prediction for high school baseball hitter using K-means clustering method", Journal of Knowledge Information Technology and Systems, 17(4), pp.733~742 (2022)
- "Tensorflow" (2019)[URL]