COMPARATIVE ANALYSIS OF MACHINE LEARNING APPROACHES FOR ASTEROID CLASSIFICATION BASED ON THE ORBITAL MOTION

SIVA ARDELIA AZZAHRA, . (2024) COMPARATIVE ANALYSIS OF MACHINE LEARNING APPROACHES FOR ASTEROID CLASSIFICATION BASED ON THE ORBITAL MOTION. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

	Text Part Covar.pdf Download (1MB)
	Text part introduction.pdf Download (1MB)
	Text part literature.pdf Restricted to Registered users only Download (4MB) | Request a copy
	Text part method.pdf Restricted to Registered users only Download (729kB) | Request a copy
	Text part result.pdf Restricted to Registered users only Download (2MB) | Request a copy
	Text part conclusion.pdf Restricted to Registered users only Download (362kB) | Request a copy
	Text part references.pdf Download (1MB)
	Text part appendix.pdf Restricted to Registered users only Download (972kB) | Request a copy

Abstract

The study determined the most appropriate machine learning classifiers to detect Potentially Hazardous Asteroid (PHA). The machine learning classifiers: K-Nearest Neighbors (KNN), Naïve Bayes, and Random Forest were applied to develop an asteroid classification program based on orbital parameters. Each classifier was evaluated by its precision, accuracy, F1-score, and recall in determining PHA and non-PHA. The Random Forest achieve the highest accuracy score at 100%, followed by the KNN classifier with an accuracy score at 97.00%, and the Naïve Bayes classifier with an accuracy score at 95.00%. The results of this research are proposed to provide a better comprehension of several machine learning classifiers performance in classifying asteroids. *****Penelitian ini menentukan metode machine learning yang paling sesuai untuk mendeteksi Potentially Hazardous Asteroid (PHA). Metode machine learning: K�Nearest Neighbors (KNN), Naïve Bayes, dan Random Forest diterapkan untuk mengembangkan program klasifikasi asteroid berdasarkan parameter orbit. Setiap metode dievaluasi berdasarkan presisi, akurasi, skor F1, dan recall dalam menentukan PHA dan non-PHA. Random Forest mencapai skor akurasi tertinggi sebesar 100%, diikuti oleh KNN dengan skor akurasi sebesar 97,00%, dan Naïve Bayes dengan skor akurasi sebesar 95,00%. Hasil penelitian ini diusulkan untuk memberikan pemahaman yang lebih baik tentang kinerja beberapa metode machine learning dalam mengklasifikasikan asteroid.

Item Type:	Thesis (Sarjana)
Additional Information:	1). Dr. Mutia Delina, M. Si 2). Dr. Janaka Adassuriya
Subjects:	Sains > Astronomi Teknologi dan Ilmu Terapan > Teknik Komputer
Divisions:	FMIPA > S1 Fisika
Depositing User:	Users 25185 not found.
Date Deposited:	22 Aug 2024 03:45
Last Modified:	22 Aug 2024 03:45
URI:	http://repository.unj.ac.id/id/eprint/50237

Actions (login required)

View Item