Image of PENERAPAN SYNTHETIC MINORITY OVERSAMPLING TECHNIQUE (SMOTE) UNTUK MENGATASI DATA IMBALANCED DALAM KLASIFIKASI KEJADIAN HUJAN MENGGUNAKAN METODE REGRESI LOGISTIK BINER
Bookmark Share

Text

PENERAPAN SYNTHETIC MINORITY OVERSAMPLING TECHNIQUE (SMOTE) UNTUK MENGATASI DATA IMBALANCED DALAM KLASIFIKASI KEJADIAN HUJAN MENGGUNAKAN METODE REGRESI LOGISTIK BINER

Classification on imbalanced data can affect the accuracy value of the classification and tends to ignore the minority class so that the prediction results will tend to the category. To overcome the problem of imbalanced data, synthetic minority oversampling technique (SMOTE) can be applied. This study aims to obtain an increase in classification accuracy by applying SMOTE to overcome imbalanced data in the classification of rain events using the binary logistic regression method. The data used is secondary data from the weather query builder dataset, namely daily data on rain events in Prabumulih City. The application of SMOTE to the binary logistic regression method for the classification of rain events resulted in an increase in the classification accuracy value in accuracy, precision and fscore, namely 0.27%, 0.91% and 0.04%, this shows that the application of SMOTE for the classification of rain events provides a better level of classification accuracy in accuracy, precision and fscore. While the level of classification accuracy in recall after the application of SMOTE decreased by 0.74%, this was caused by overfitting of synthetic data generated by SMOTE in the non-rain class which could make the model too focused on recognizing train data patterns, thus losing the ability to recognize patterns in test data.

Availability
#
Central Library (Reference) T1693422025
T169342
Available but not for loan - Not for Loan
Detail Information
Series Title
-
Call Number
T1693422025
Publisher
: Prodi Ilmu Matematika, Fakultas Matematika Dan Ilmu Pengetahuan Alam Universitas Sriwijaya.,
Collation
xiii, K-6 hlm. : ilus. ; tab. ; 29 cm
Language
Indonesia
ISBN/ISSN
-
Classification
510.07
Content Type
-
Media Type
-
Carrier Type
-
Edition
-
Subject(s)
Matematika
Specific Detail Info
-
Statement of Responsibility
Other version/related
File Attachment
Comments
You must be logged in to post a comment