Ensemble Machine Learning Models for Simulating the Missile Defense System
- Authors
- Jin, Sihwa; Dahouda, Mwamba Kasongo; Joe, Inwhee
- Issue Date
- Jan-2023
- Publisher
- Springer Science and Business Media Deutschland GmbH
- Keywords
- LGBM; Machine learning; Missile; Random forest; Simulator; XGBoost
- Citation
- Lecture Notes in Networks and Systems, v.597 LNNS, pp.142 - 156
- Indexed
- SCOPUS
- Journal Title
- Lecture Notes in Networks and Systems
- Volume
- 597 LNNS
- Start Page
- 142
- End Page
- 156
- URI
- https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/182536
- DOI
- 10.1007/978-3-031-21438-7_12
- ISSN
- 2367-3370
- Abstract
- This paper simulated the missile engagement situation using a simulator and conducted a machine learning study based on the generated data. The simulator simulates missile engagements between the enemy and our forces and collects data. The collected data is learned using random forest, XGBoost, and LGBM models after preprocessing. In addition, hyperparameter adjustments were performed for each model to find the optimal parameters. Different metrics for accuracy, F1-score, and ROC-AUC were used for performance comparison. As a result of the experiment, XGBoost showed the best performance in performance indicators, and LGBM was the fastest in terms of learning speed. This paper suggests that XGBoost, which is slow in learning speed but has the best accuracy and performance indicators, is suitable for one-to-one interception situations, and LGBM, which is fast in learning and has excellent performance indicators, is suitable for many-to-many interception situations.
- Files in This Item
-
Go to Link
- Appears in
Collections - 서울 공과대학 > 서울 컴퓨터소프트웨어학부 > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.