Text

EVALUASI KINERJA MODEL KLASIFIKASI KEJADIAN HUJAN KABUPATEN OGAN ILIR DENGAN METODE REGRESI LOGISTIK BINER DAN K-NEAREST NEIGHBOR SEBELUM DAN SESUDAH PENERAPAN SYNTHETHIC MINORITY OVER-SAMPLING TECHNIQUE

---

Rainfall is a key component of the water cycle and plays an important role in the ecosystem, especially in tropical regions such as Indralaya Regency. Rain affects various operational aspects in the field, particularly in agriculture and plantations. Binary Logistic Regression and K-Nearest Neighbor (KNN) can be used to classify rainfall events. One common issue in rainfall event classification is class imbalance. To address this problem, a class-balancing technique called Synthetic Minority Over-Sampling Technique (SMOTE) is used, which helps balance the data by generating synthetic samples for the minority class. This study aims to obtain accuracy, precision, recall, and f-score values for Binary Logistic Regression and KNN both before and after applying SMOTE, and to compare model performance to identify the best model. The data used are secondary rainfall event data for Ogan Ilir Regency from 2018 to 2023, obtained from the visualcrossing.com website, with 14 independent variables and 1 dependent variable. The results show that Binary Logistic Regression after SMOTE provides the best performance with an accuracy of 83.56%, precision of 87.97%, recall of 81.44%, and f-score of 84.58%. Other results are Binary Logistic Regression before SMOTE (accuracy 79.18%, precision 74.14%, recall 95.79%, f-score 83.58%), KNN before SMOTE (accuracy 78.77%, precision 77.73%, recall 86.39%, f-score 81.83%), and KNN after SMOTE (accuracy 74.38%, precision 80.73%, recall 70.54%, f-score 75.29%). Keywords: Rainfall, SMOTE, Binary Logistic Regression, KNN

---

Availability

Detail Information

Series Title

-

Call Number

T1843342025

Publisher

Indralaya : Prodi Ilmu Matematika, Fakultas Matematika Dan Ilmu Pengetahuan Alam Universitas Sriwijaya., 2025

Collation

x, 49 hlm.; ilus.; tab.; 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

519.507

Content Type

Text

Media Type

unmediated

Carrier Type

other (computer)

Edition

-

Subject(s)

Prodi Ilmu Matematika
Matematika Statistikal

Specific Detail Info

-

Statement of Responsibility

SEW

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment