Skripsi

IMPLEMENTASI KNN-SMOTE-ROS-RUS DALAM MENGKLASIFIKASIKAN KEJADIAN HUJAN MENGGUNAKAN METODE NAÏVE BAYES

---

Topographically, Palembang city consists mostly of swamps and lowlands. The city's condition of being surrounded by rivers and its moderate annual rainfall of 2500-3000 mm makes Palembang prone to flooding. Therefore, accurately predicting rainfall events is crucial to minimize the impact on daily activities. The rainfall prediction classification process, missing data and imbalanced data were encountered in the rainfall event dataset. Missing data was addressed using KNN (K=17) and standardization was performed to standardize the data scale. Subsequently, to address imbalanced data, SMOTE, ROS and RUS techniques were applied. Thus, for the imbalanced data, an accuracy of 98.77%, precision of 97.83%, and recall of 100% were obtained. Furthermore, for the balanced data using SMOTE, an accuracy of 98.22%, precision of 96.89%, and recall of 100% were achieved, while ROS yielded an accuracy of 98.90%, precision of 98.06%, and recall of 100%, and RUS yielded an accuracy of 99.31%, precision of 98.78%, and recall of 100%. Based on these findings, the best rainfall event prediction classification for Palembang city was obtained by addressing imbalanced data using the RUS technique.

---

Availability

Detail Information

Series Title

-

Call Number

T1461722024

Publisher

Indralaya : Prodi Ilmu Matematika, Fakultas Matematika Dan Ilmu Pengetahuan Alam Universitas Sriwijaya., 2024

Collation

xiv, 50 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