DATA MINING USING THE GRADIENT BOOSTED TREES METHOD TO CLASSIFY THE TYPES OF TRAFFIC VIOLATIONS IN TULUNGAGUNG

Wirasaputra, Dewa Eka (2026) DATA MINING USING THE GRADIENT BOOSTED TREES METHOD TO CLASSIFY THE TYPES OF TRAFFIC VIOLATIONS IN TULUNGAGUNG. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 19081010139.-cover.pdf Download (702kB)
	Text (Bab 1) 19081010139.-bab1.pdf Download (229kB)
	Text (Bab 2) 19081010139.-bab2.pdf Restricted to Repository staff only until 22 July 2029. Download (581kB)
	Text (Bab 3) 19081010139.-bab3.pdf Restricted to Repository staff only until 22 July 2029. Download (603kB)
	Text (Bab 4) 19081010139.-bab4.pdf Restricted to Repository staff only until 22 July 2029. Download (1MB)
	Text (Bab 5) 19081010139.-bab5.pdf Download (218kB)
	Text (Daftar Pustaka) 19081010139.-daftarpustaka.pdf Download (208kB)

Abstract

This study aims to apply the Gradient Boosted Trees (GBT) method to classify types of traffic violations in Tulungagung Regency into three categories: minor, moderate, and severe. The dataset used consists of 8,621 records from Sistem Informasi Penelusuran Perkara (SIPP) of the Tulungagung District Court for the year 2023. The research stages included data preprocessing comprising data cleaning, data transformation, feature selection, and data labeling followed by the development of a classification model using the violation article, vehicle type, and type of evidence as features. Experiments were conducted using data split scenarios of 70:30, 75:25, and 80:20 data split scenarios, as well as testing variations in the hyperparameters n_estimators (100, 200, and 300), max_depth (3, 5, and 7), learning_rate (0.01, 0.05, and 0.1), and subsample (0.8, 0.9, and 1.0). The test results show that the 80:20 data split scenario produced the best performance compared to the other scenarios. The best parameter combination was obtained with n_estimators = 200, max_depth = 3, learning_rate = 0.05, and subsample = 0.9. The best model obtained through GridSearchCV optimization yielded an accuracy of 64.54%, precision of 64.24%, recall of 46,24%, and an F1-score of 47,53%. The results of this study indicate that the Gradient Boosted Trees method can be used to classify types of traffic violations and improve the effectiveness of systematic traffic violation data analysis.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Via, Yisti Vita	NIDN0025048602	yistivia.if@upnjatim.ac.id
Thesis advisor	Al Haromainy, Muhammad Muharrom	NIDN0701069503	muhammad.muharrom.if@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics > QA75 Electronic computers. Computer science

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Dewa Eka Wirasaputra

Date Deposited:

22 Jul 2026 07:00

Last Modified:

22 Jul 2026 08:01

URI:

https://repository.upnjatim.ac.id/id/eprint/57372