Prediksi Hasil Pertandingan Mobile Legends Professional League (MPL) Berdasarkan Statistik Individu Pemain Menggunakan CatBoost

Ardine, Ciptaagung Firjat (2025) Prediksi Hasil Pertandingan Mobile Legends Professional League (MPL) Berdasarkan Statistik Individu Pemain Menggunakan CatBoost. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 21081010164-cover.pdf Download (657kB)
	Text (Bab 1) 21081010164-bab1.pdf Download (254kB)
	Text (Bab 2) 21081010164-bab2.pdf Restricted to Repository staff only until 5 December 2027. Download (1MB)
	Text (Bab 3) 21081010164-bab3.pdf Restricted to Repository staff only until 5 December 2027. Download (1MB)
	Text (Bab 4) 21081010164-bab4.pdf Restricted to Repository staff only until 5 December 2027. Download (3MB)
	Text (Bab 5) 21081010164-bab5.pdf Download (77kB)
	Text (Daftar Pustaka) 21081010164-daftarpustaka.pdf Download (235kB)
	Text (Lampiran) 21081010164-lampiran.pdf Restricted to Repository staff only Download (487kB)

Abstract

Mobile Legends: Bang Bang (MLBB) is one of the most popular mobile-based MOBA esports games, with the Mobile Legends Professional League (MPL) as its main league. Predicting the match winner is one of the most popular things to do. Player performance statistics are one of the key factor that affects team results. Machine learning can be used to learn patterns from player statistics and give win probabilities for the team. Categorical Boosting (CatBoost) is gradient boosting algorithm that increasingly used because of its good performance. The goals of this research is to develop a predictive model for MPL match outcome using CatBoost. The dataset is individual player statistics from MPL Malaysia Season 14 matches. CatBoost models were tested using 5-Fold Cross Validation. The results shows CatBoost default model achieved 96.15% average accuracy score with the 26.08 seconds computation duration. Grid Search hyperparameter tuning increase the model performance with 96.50% average accuracy score and 2.45 seconds computation duration. Shapley Additive exPlanations feature importance analysis shows Gold Per Minute (GPM), Tower Damage, and KDA (Kill, Death, Assist) are the most important features. The results when model was tested using different seasons and regions of MPL shows good performance and implementation potential

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Mandyartha, Eka Prakarsa	NIDN0725058805	eka_prakarsa.fik@upnjatim.ac.id
Thesis advisor	Junaidi, Achmad	NIDN0710117803	achmadjunaidi.if@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Ciptaagung Firjat Ardine

Date Deposited:

08 Dec 2025 01:46

Last Modified:

08 Dec 2025 02:12

URI:

https://repository.upnjatim.ac.id/id/eprint/48103