Analisis Perbandingan Metode Decision Tree, Naive Bayes, dan Radial Basis Function Neural Network pada Klasifikasi Tinggi Muka Air Laut Marina Ancol

Nadhifah, Qatrunnada Gusti (2023) Analisis Perbandingan Metode Decision Tree, Naive Bayes, dan Radial Basis Function Neural Network pada Klasifikasi Tinggi Muka Air Laut Marina Ancol. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 06311940000033-Undergraduate_Thesis.pdf] Text
06311940000033-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2025.

Download (3MB) | Request a copy

Abstract

Secara geografis, Indonesia merupakan negara kepulauan dengan luas lautan yang mencapai dua per tiga dibandingkan dengan luas daratannya. Berdasarkan hal tersebut, salah satu potensi yang dapat dikembangkan pada bidang perekonomian Indonesia adalah wisata bahari. Pantai Marina Ancol merupakan salah satu kawasan wisata bahari yang terletak di wilayah Jakarta Utara. Untuk menciptakan rasa keamanan dan kenyamanan bagi para wisatawan Pantai Marina Ancol, analisis klasifikasi tinggi muka air laut perlu dilakukan sebagai upaya antisipasi terhadap terjadinya bencana banjir rob yang merupakan peristiwa penggenangan daratan akibat terjadinya kenaikan permukaan air laut. Pada penelitian ini akan dilakukan analisis klasifikasi tinggi muka air laut Marina Ancol menggunakan metode Decision Tree, Naïve Bayes, dan Radial Basis Function Neural Network untuk mengetahui metode terbaik yang dapat digunakan dalam mengklasifikasikan tinggi muka air ke dalam empat kategori muka air yaitu Siaga 1, Siaga 2, Siaga 3, dan Siaga 4 berdasarkan accuracy score dan false negative rate. Dalam memprediksikan kategori muka air laut Marina Ancol, model yang didapat dengan metode Decision Tree menghasilkan nilai akurasi sebesar 85,19%, nilai false negative rate Siaga 1 sebesar 16,67%, dan nilai false negative rate Siaga 2 sebesar 12,24%. Dengan menggunakan metode Naïve Bayes, nilai akurasi yang dihasilkan dari model adalah sebesar 73,37%, nilai false negative rate Siaga 1 sebesar 16,67%, dan nilai false negative rate Siaga 2 sebesar 24,27%. Arsitektur model Radial Basis Function Neural Network yang terbaik adalah model dengan jumlah input layer sebanyak tiga, hidden neuron sebanyak lima belas, dan output neuron sebanyak empat. Nilai akurasi model RBFNN (3-15-4) menunjukkan nilai sebesar 86,64% nilai false negative rate Siaga 1 sebesar 100%, dan nilai false negative rate Siaga 2 sebesar 14,73%. Berdasarkan hasil tersebut, Decision Tree merupakan metode yang menghasilkan model lebih baik dibandingkan Radial Basis Function Neural Network dan Naïve Bayes dalam melakukan prediksi kategori muka air laut Marina Ancol.
================================================================================================================================
Geographically, Indonesia is an archipelagic country with an ocean area that reaches two-thirds compared to its land area. Based on this, one of the potentials that can be developed in the Indonesian economy sector is marine tourism. Marina Ancol Beach is one of the marine tourism areas located in the North Jakarta area. To create a sense of security and comfort for tourists of Ancol Marina Beach, an analysis of the classification of sea level height needs to be carried out as an effort to anticipate the occurrence of tidal flood disasters which is land flooding events due to sea level rise. In this study, a study comparison analysis of the classification of water level height on Marina Ancol will be carried out using the Decision Tree, Naïve Bayes, and Radial Basis Function Neural Network methods to find out the best method that can be used in classifying water level height into four water level categories, namely Siaga 1, Siaga 2, Siaga 3, and Siaga 4 based on the accuracy score and false negative rate. In predicting the Marina Ancol sea level category, the model obtained by the Decision Tree method produced an accuracy value of 85.19%, the value of Siaga 1 false negative rate is 16.67%, and the Siaga 2 false negative rate value is 12.24%. Using the Naïve Bayes method, the accuracy value generated from the model is 73.37%, the false negative rate value of Siaga 1 is 16.67%, and the Siaga 2 false negative rate value is 24.27%. The best Radial Basis Function Neural Network model architecture is a model with three input layers, fifteen hidden neurons, and four output neurons. The accuracy value of the RBFNN model (3-15-4) shows a value of 86.64%, the value of Siaga 1 false negative rate is 100%, and the Siaga 2 false negative rate value is 14.73%. Based on these results, Decision Tree is a method that produces better models than Radial Basis Function Neural Network and Naïve Bayes in predicting Marina Ancol sea water level categories.

Item Type: Thesis (Other)
Uncontrolled Keywords: Classification, Decision Tree, Naïve Bayes, Radial Basis Function Neural Network, Water Level, Klasifikasi, Tinggi Muka Air
Subjects: Q Science > QA Mathematics > QA279.5 Bayesian statistical decision theory.
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User: Qatrunnada Gusti Nadhifah
Date Deposited: 13 Jul 2023 01:31
Last Modified: 13 Jul 2023 01:31
URI: http://repository.its.ac.id/id/eprint/98439

Actions (login required)

View Item View Item