Perbandingan Optimasi Algoritma Random Forest Menggunakan Random Search dan Grid Search pada Klasifikasi Kualitas Air Bersih

Firmansyah, Adi Fajri (2025) Perbandingan Optimasi Algoritma Random Forest Menggunakan Random Search dan Grid Search pada Klasifikasi Kualitas Air Bersih. Undergraduate thesis, UPN VETERAN JAWA TIMUR.

[img] Text (Cover)
21081010175-cover.pdf

Download (3MB)
[img] Text (Bab 1)
21081010175-bab1.pdf

Download (88kB)
[img] Text (Bab 2)
21081010175-bab2.pdf
Restricted to Repository staff only until 24 July 2027.

Download (333kB)
[img] Text (Bab 3)
21081010175-bab3.pdf
Restricted to Repository staff only until 24 July 2027.

Download (1MB)
[img] Text (Bab 4)
21081010175-bab4.pdf
Restricted to Repository staff only until 24 July 2027.

Download (803kB)
[img] Text (Bab 5)
21081010175-bab5.pdf

Download (15kB)
[img] Text (Daftar Pustaka)
21081010175-daftarpustaka.pdf

Download (150kB)
[img] Text (Lampiran)
21081010175-lampiran.pdf
Restricted to Repository staff only until 24 July 2027.

Download (490kB)

Abstract

Potable water suitable for consumption is a fundamental human need that plays a crucial role in health. In Indonesia, challenges regarding water quality and accessibility remain significant due to infrastructure limitations and environmental pollution. This study aims to compare the effectiveness of Random Forest algorithm optimization using Random Search and Grid Search in clean water quality classification. The dataset used is “Water Quality and Potability” obtained from the Kaggle page, consisting of 3,276 data with 9 parameter features and 1 target variable. The entire classification process is carried out on the Google Colab platform. Evaluation results show that the default Random Forest model achieved an accuracy of 84.12% with a computation time of 3 seconds. The Random Search-optimized model reached 84.38% accuracy in 2 hours 2 minutes and 54 seconds, while the Grid Search-optimized model achieved the highest accuracy of 85.62% in 5 hours 12 minutes. Overall, hyperparameter optimization significantly improved classification performance, particularly with Grid Search. However, this improvement was accompanied by a substantial increase in computation time, which should be considered when selecting optimization methods.

Item Type: Thesis (Undergraduate)
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorRahmat, BasukiNIDN0023076907basukirahmat.if@upnjatim.ac.id
Thesis advisorHaromainy, Muhammad Muharrom AlNIDN0701069503muhammad.muharrom.if@upnjatim.ac.id
Subjects: T Technology > T Technology (General)
Divisions: Faculty of Computer Science > Departemen of Informatics
Depositing User: Adi Fajri F
Date Deposited: 25 Jul 2025 02:19
Last Modified: 25 Jul 2025 02:19
URI: https://repository.upnjatim.ac.id/id/eprint/40815

Actions (login required)

View Item View Item