KLASIFIKASI MUTU AIR SUNGAI INDONESIA MENGGUNAKAN REGRESI LOGISTIK ORDINAL BERBASIS ADAPTIVE SYNTHETIC SAMPLING

ERIN NAUDY KEMALASARI, . (2026) KLASIFIKASI MUTU AIR SUNGAI INDONESIA MENGGUNAKAN REGRESI LOGISTIK ORDINAL BERBASIS ADAPTIVE SYNTHETIC SAMPLING. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

	Text Cover.pdf Download (2MB)
	Text BAB I.pdf Download (1MB)
	Text BAB II.pdf Restricted to Registered users only Download (525kB) | Request a copy
	Text BAB III.pdf Restricted to Registered users only Download (350kB) | Request a copy
	Text BAB IV.pdf Restricted to Registered users only Download (395kB) | Request a copy
	Text BAB V.pdf Restricted to Registered users only Download (228kB) | Request a copy
	Text Daftar Pustaka.pdf Download (244kB)
	Text Lampiran.pdf Restricted to Registered users only Download (1MB) | Request a copy

Abstract

Mutu air sungai merupakan indikator penting dalam pengelolaan lingkungan. Di Indonesia, status mutu air sungai ditentukan menggunakan Indeks Pencemaran yang mengelompokkan kualitas air menjadi empat kategori, yaitu memenuhi baku mutu, cemar ringan, cemar sedang, dan cemar berat. Ketidakseimbangan jumlah data pada setiap kategori dapat menyebabkan model klasifikasi cenderung memihak kelas mayoritas. Penelitian ini bertujuan menerapkan regresi logistik ordinal untuk mengklasifikasikan status mutu air sungai di Indonesia serta mengevaluasi pengaruh Adaptive Synthetic Sampling (ADASYN) dalam meningkatkan performa klasifikasi pada data tidak seimbang dengan membandingkannya terhadap model tanpa penyeimbangan data dan Synthetic Minority Over-sampling Technique (SMOTE). Data yang digunakan merupakan data kualitas air sungai dari Kementerian Lingkungan Hidup dan Kehutanan dengan variabel prediktor amonia, Biological Oxygen Demand (BOD), Chemical Oxygen Demand (COD), Dissolved Oxygen (DO), dan Nitrat. Evaluasi model dilakukan menggunakan confusion matrix berdasarkan nilai Accuracy, Precision, Recall, dan F-Measure. Hasil penelitian menunjukkan bahwa amonia, BOD, COD, dan Nitrat berpengaruh signifikan terhadap status mutu air sungai, sedangkan DO tidak berpengaruh signifikan. Model tanpa penyeimbangan data menghasilkan Accuracy tertinggi, sedangkan model berbasis ADASYN menghasilkan Macro Precision dan Macro F-Measure tertinggi sehingga memberikan performa klasifikasi yang lebih seimbang dibandingkan model tanpa penyeimbangan data maupun SMOTE.****River water quality is a key indicator in environmental management. In Indonesia, water quality status is determined using a Pollution Index that classifies water quality into four categories: meeting quality standards, slightly polluted, moderately polluted, and heavily polluted. An imbalance in the amount of data across each category can cause classification models to favor the majority class. This study aims to apply ordinal logistic regression to classify water quality status in Indonesia and to evaluate the effect of Adaptive Synthetic Sampling (ADASYN) on improving classification performance for imbalanced data by comparing it to a model without data balancing and the Synthetic Minority Over-sampling Technique (SMOTE). The data used consists of river water quality data from the Ministry of Environment and Forestry, with the following predictor variables: ammonia, Biological Oxygen Demand (BOD), Chemical Oxygen Demand (COD), Dissolved Oxygen (DO), and Nitrat. Model evaluation was conducted using a confusion matrix based on Accuracy, Precision, Recall, and F-Measure. The results show that ammonia, BOD, COD, and Nitrat have a significant effect on water quality status, while DO does not have a significant effect. The model without data balancing produced the highest Accuracy, while the ADASYN-based model produced the highest Macro Precision and Macro F-Measure, thus providing more balanced classification performance compared to both the model without data balancing and the SMOTE model.

Item Type:	Thesis (Sarjana)
Additional Information:	1). Dr. Vera Maya Santi, S.Si., M.Si. ; 2). Qorry Meidianingsih, S.Si., M.Si.
Subjects:	Ilmu Sosial > Statistik Sains > Sains, Ilmu Pengetahuan Alam
Divisions:	FMIPA > S1 Statistika
Depositing User:	Erin Naudy Kemalasari .
Date Deposited:	20 Aug 2026 07:32
Last Modified:	20 Aug 2026 07:32
URI:	http://repository.unj.ac.id/id/eprint/72221

Actions (login required)

View Item