Image of ANALISIS KLASIFIKASI KEJADIAN CUACA HUJAN DI KOTA LUBUK LINGGAU MENGGUNAKAN METODE NAIVE BAYES DAN LOGISTIC REGRESSION DENGAN DAN TANPA SYNTHETIC MINORITY OVER-SAMPLING (SMOTE)
Bookmark Share

Skripsi

ANALISIS KLASIFIKASI KEJADIAN CUACA HUJAN DI KOTA LUBUK LINGGAU MENGGUNAKAN METODE NAIVE BAYES DAN LOGISTIC REGRESSION DENGAN DAN TANPA SYNTHETIC MINORITY OVER-SAMPLING (SMOTE)

Weather is a dynamic atmospheric phenomenon that influences various sectors, particularly in tropical regions with complex climatic characteristics such as Lubuk Linggau City. This study aims to classify weather events using the Naïve Bayes and Logistic Regression methods, both without and with the application of the Synthetic Minority Over-sampling Technique (SMOTE), and to determine the most effective method for weather classification. The dataset consists of 3,000 daily weather observations obtained from Visual Crossing, comprising 13 independent variables and one dependent variable representing weather conditions. The classification process was conducted by dividing the dataset into training and testing sets and evaluated using a confusion matrix along with performance metrics including Accuracy, Precision, Recall, and F1-Score. The results indicate that both methods achieve very high classification performance. Naïve Bayes achieved an accuracy of 99.56% without SMOTE and 99.44% with SMOTE, while Logistic Regression obtained 100% accuracy without SMOTE and 99.78% with SMOTE. The application of SMOTE did not significantly improve the performance of either model, indicating that the original dataset already contains sufficiently strong patterns for classification. Based on the comparative results, Logistic Regression demonstrates more stable performance and is identified as the most effective method for classifying weather events in Lubuk Linggau City. The findings of this study are expected to support the development of weather classification systems that can assist risk mitigation and data-driven decision making at the regional level.

Availability
#
Central Library (Reference) T1949842026
T194984
Available but not for loan - Not for Loan
Detail Information
Series Title
-
Call Number
T1949842026
Publisher
Indralaya : Prodi Ilmu Matematika, Fakultas Matematika Dan Ilmu Pengetahuan Alam Universitas Sriwijaya.,
Collation
xvi, 91 hlm.; ilus.; 29 cm
Language
Indonesia
ISBN/ISSN
-
Classification
006.315 07
Content Type
Text
Media Type
unmediated
Carrier Type
-
Edition
-
Subject(s)
Curah hujan
Prodi Ilmu Matematika
Specific Detail Info
-
Statement of Responsibility
Other version/related
File Attachment
Comments
You must be logged in to post a comment