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

ADININGRUM, SYAKIRA and Resti, Yulia and Dewi, Novi Rustiana (2024) IMPLEMENTASI KNN-SMOTE-ROS-RUS DALAM MENGKLASIFIKASIKAN KEJADIAN HUJAN MENGGUNAKAN METODE NAÏVE BAYES. Undergraduate thesis, Sriwijaya University.

	Text RAMA_44201_08011282025035.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_44201_08011282025035_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_44201_08011282025035_0019077302_0013117004_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_44201_08011282025035_0019077302_0013117004_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (412kB) | Request a copy
	Text RAMA_44201_08011282025035_0019077302_0013117004_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (370kB) | Request a copy
	Text RAMA_44201_08011282025035_0019077302_0013117004_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (788kB) | Request a copy
	Text RAMA_44201_08011282025035_0019077302_0013117004_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (202kB) | Request a copy
	Text RAMA_44201_08011282025035_0019077302_0013117004_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (325kB) | Request a copy
	Text RAMA_44201_08011282025035_0019077302_0013117004_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (237kB) | Request a copy

Abstract

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.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	KNN, SMOTE, ROS, RUS, Hujan
Subjects:	Q Science > QA Mathematics > QA75-76.95 Calculating machines > QA76.9.B45 Big data. Machine learning. Quantitative research. Metaheuristics.
Divisions:	08-Faculty of Mathematics and Natural Science > 44201-Mathematics (S1)
Depositing User:	Syakira Adiningrum
Date Deposited:	04 Jun 2024 01:05
Last Modified:	04 Jun 2024 01:05
URI:	http://repository.unsri.ac.id/id/eprint/146172

Actions (login required)

View Item