Peramalan Jumlah Kasus Demam Berdarah di Kabupaten Malang Menggunakan Metode Radial Basis Function Neural Network - Genetic Algorithm

Ramadhani, Humayyun Nabila (2020) Peramalan Jumlah Kasus Demam Berdarah di Kabupaten Malang Menggunakan Metode Radial Basis Function Neural Network - Genetic Algorithm. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of ACCESS CLOSED BY AUTHOR] Text (ACCESS CLOSED BY AUTHOR)
05211640000008-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only

Download (2MB) | Request a copy

Abstract

Demam Berdarah Dengue (DBD) merupakan penyakit yang ditularkan ke manusia melalui gigitan nyamuk Aedes yang terinfeksi virus Dengue. Penyakit ini sudah menyebar luas ke seluruh Indonesia termasuk Kabupaten Malang. Pemerintah dan Dinas Kesehatan di Kabupaten Malang telah melakukan berbagai upaya seperti pencegahan dan sosialisasi, namun jumlah kasus demam berdarah masih mengalami peningkatan sebesar 9,53% tiap tahun dengan jumlah kematian juga mengalami pengingkatan sebesar 5% per tahun. Di Kabupaten Malang dengan adanya tiga pembagian wilayah dataran di dalamnya, yaitu dataran rendah, dataran sedang, dan dataran tinggi, terdapat banyak variabel-variabel yang dapat mempengaruhi terus meningkatnya angka kasus demam berdarah.

Peramalan jumlah kasus Demam Berdarah pada penelitian ini menggunakan metode Radial Basis Function Neural Network. Selain hal tersebut, dilakukan pula pendekatan Algortima Genetika untuk penentuan parameter yang paling optimal agar didapatkan hasil akurasi yang optimal dan minim kesalahan. Sementara jenis variabel yang digunakan dalam meramalkan jumlah kasus demam berdarah di Kabupaten Malang, yaitu variabel suhu, curah hujan, kelembapan, dan jumlah penduduk.

Hasil pengukuran akurasi antara metode Radial Basis Function Neural Network dengan metode hybrid Radial Basis Function Neural Network – Genetic Algorithm menunjukkan bahwa metode Radial Basis Function Neural Network merupakan metode yang dapat memberikan hasil peramalan lebih baik sebesar 41.44 % daripada metode hybrid Radial Basis Function Neural Network – Genetic Algorithm. Hasil nilai Mean Square Error data pengujian paling minimum yaitu sebesar 3.625 pada Kecamatan Ngajum dan Symmetric Mean Absolute Percentage Error paling minimum yaitu sebesar 39.427 % pada Kecamatan Pakisaji dengan diketahui variabel-variabel independen yang paling berpengaruh adalah variabel curah hujan pada dataran rendah dan dataran sedang.
==================================================================================================================================
Dengue Fever is a disease that is transmitted to humans through the bite of Aedes mosquitoes infected with Dengue virus. This disease has spread widely throughout Indonesia, including Malang Regency. The Government and the Department of Health in Malang Regency have made various efforts such as prevention and outreach, but the number of dengue fever cases has increased by 9.53% per year with the number of deaths also increasing by 5% per year. In Malang Regency with three divisions in the plain area, namely lowlands, mediumlands, and highlands, there are many specific variables based on the results of previous studies which are variables that can influence the increasing number of dengue fever cases.

Forecasting the number of Dengue Fever in this research using the Radial Basis Function Neural Network method. Besides this, the Algortima Genetics approach is also carried out to determine the most optimal parameters in order to obtain optimal accuracy and minimal error results. While the types of variables used in predicting the number of dengue cases in Malang Regency are temperature, rainfall, humidity, and population

The results of the accuracy measurement between the Radial Basis Function Neural Network method and the Hybrid Radial Basis Function Neural Network - Genetic Algorithm method show that the Radial Basis Function Neural Network method is a method that can provide better forecasting results by 41.44% than the hybrid Radial Basis Function Neural Network method - Genetic Algorithm. The result of the Mean Square Error value of the minimum testing data is 3,625 in the Ngajum District and the minimum minimum Symmetric Mean Absolute Percentage Error is 39,427% in the Pakisaji District with the most influential independent variables being the rainfall variable in the lowlands and temperate plains.

Item Type: Thesis (Other)
Additional Information: 3100020084860 RSSI 006.3 Ram p-1
Uncontrolled Keywords: Berdarah Dengue, Algoritma Genetika, Radial Basis Function Neural Network, Optimasi, Parameter, Nilai Akurasi, Mean Square Error, Symmetric Mean Absolute Percentage
Subjects: Q Science > QA Mathematics > QA402.5 Genetic algorithms.
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis
Depositing User: HUMAYYUN NABILA RAMADHANI
Date Deposited: 05 Jan 2023 07:28
Last Modified: 05 Jan 2023 07:28
URI: http://repository.its.ac.id/id/eprint/72760

Actions (login required)

View Item View Item