Skripsi

PENERAPAN SYNTHETIC MINORITY OVER-SAMPLING TECHNIQUE PADA PENGKLASIFIKASIAN KEJADIAN HUJAN KOTA PRABUMULIH MENGGUNAKAN METODE FUZZY NAÏVE BAYES DAN K-NEAREST NEIGHBOR

---

Penilaian

---

Imbalanced class data can affect classification performance. An imbalanced class occurs when the minority class is less than the majority class. Imbalanced class can be overcome by resampling one of them using synthetic minority over-sampling technique (SMOTE). This research aims to balance the class by applying SMOTE to the Prabumulih city rainfall event classification using the Fuzzy Naïve Bayes and K Nearest Neighbor methods. The data used is rainfall event data in Prabumulih city sourced from the visual crossing website with 17 variables, totalling 2556 data. Train data is 71,44% or 1826 data and test data is 28,56% or 730 data. The accuracy of rainfall event classification before the application of SMOTE using Fuzzy Naïve Bayes method resulted in accuracy, precision, recall, and f-score of 73,42%, 71,43%, 65,63%, and 68,40%, respectively. After the application of SMOTE using the Fuzzy Naïve Bayes method, the accuracy and precision decreased by 2,6% and 7,46%. In contrast, recall and f-score increased by 10,93% and 1,3%. Meanwhile, before the application of SMOTE using the K-Nearest Neighbor method, the accuracy, precision, recall, and f-score were 69,45%, 75,66%, 44,68%, and 69,93%. After the application of SMOTE using the K-Nearest Neighbor method, the accuracy, recall, and f-score increased by 3,84%, 26,57%, and 0,12%, respectively. On the other hand, precision decreased by 6,78%. The application of SMOTE using the K-Nearest Neighbor method has a better classification accuracy, seen from the accuracy, recall, and f-score values that increase compared to the application of SMOTE using the Fuzzy Naïve Bayes method.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2407003892	2407003892	T1461702024	Central Library (references)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T1461702024

Publisher

Indralaya : Jurusan Matematika Fakultas Matematika dan Ilmu Pengetahuan Alam Universitas Sriwijaya., 2024

Collation

xii, 65 hlm.; ilus.; tab.; 28 cm

Language

Indonesia

ISBN/ISSN

-

Classification

519.807

Content Type

Text

Media Type

unmediated

Carrier Type

unspecified

Edition

-

Subject(s)

Jurusan Matematika
imbalanced class

Specific Detail Info

-

Statement of Responsibility

UIN Dimas

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment