Deradjat, Andy Kusuma (2022) Analisis Dan Pemetaan Kerawanan Covid-19 Menggunakan Metode Machine Learning Technique Random Forest (Studi Kasus: Unit Pengembangan Ii Kertajaya Dan Iii Tambak Wedi Kota Surabaya). Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 03311840000042-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
03311840000042-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (6MB) |
Abstract
Pada bulan Maret 2020 pandemi COVID-19 telah melanda seluruh dunia termasuk Indonesia. Analisis dan pemetaan kerawanan sangat dibutuhkan untuk mengembangkan informasi terkait pandemi COVID-19 dari beberapa faktor yang yang mempengaruhi penularan COVID-19, salah satunya adalah faktor sosial-ekonomi yang memiliki tingkat interaksi antar manusia yang cukup beresiko. Dengan Machine Learning Technique Random forest dapat membantu dalam memberikan keputusan tingkatan atau prioritas parameter dalam analisis dan pemetaan kerawanan pandemi COVID-19 dengan akurat, mendukung keputusan, dan tindakan pencegahan. Penelitian ini mencoba untuk melakukan analisis dan pemetaan kerawanan pandemi COVID-19 berdasarkan faktor sosial-ekonomi di Unit Pengembangan II Kertajaya dan III Tambak Wedi Kota Surabaya. Penelitian ini menggunakan data statistik jumlah kasus penderita COVID-19 pada bulan Maret tahun 2020 hingga Agustus tahun 2021 dan data fasilitas sosial-ekonomi berupa jumlah fasilitas kesehatan (rumah sakit, puskesmas, dan klinik), objek wisata (tempat wisata dan taman), perbankan (atm dan bank), pasar, mall, SPBU, dan terminal. Hasil pengolahan random forest didapatkan tingkat pengaruh dari setiap parameter terhadap penyebaran COVID-19, yaitu fasilitas kesehatan 24,135%, terminal 20,338%, objek wisata 19,916%, mall 19,574%, pasar 11,317%, perbankan 2,628%, dan SPBU 2,092%. Berdasarkan penelitian terdahulu, uji akurasi model yang dihasilkan dapat dikatakan baik dengan nilai akurasi sebesar 0,946, nilai kappa sebesar 0,892, dan nilai AUC sebesar 0,984. Hasil pemetaan dari model random forest tersebut, didapatkan daerah dengan tingkat kerawanan rendah pada bagian sebelah utara Kecamatan Kenjeran dan sebelah timur Kecamatan Sukolilo dan Kecamatan Mulyorejo, sedangkan tingkat kerawanan tinggi pada bagian tengah Kecamatan Mulyorejo dan Kecamatan Sukolilo. Berdasarkan hasil tersebut, diharapkan pemerintah dapat memaksimalkan program pembatasan sosial dengan fokusan konsentrasi wilayah menurut tingkat kerawanan agar pengendalian pandemi COVID-19 dapat sesuai sasaran. Selain itu, masyarakat juga harus ikut berpartisipasi mendukung program yang diterapkan oleh pemerintah dengan menjaga interaksi sosial terutama pada fasilitas sosial-ekonomi yang merupakan sumber penyebaran COVID-19 yang cukup beresiko terhadap penyebaran pandemi COVID-19.
===================================================================================================================================
In March 2020 the COVID-19 pandemic had hit the whole world, including Indonesia. Susceptibility analysis is urgently needed to develop information related to the COVID-19 pandemic from several factors that influence the transmission of COVID-19, one of which is socio-economic factors that have a fairly risky level of human interaction. With Machine Learning Technique Random forest can assist in making decisions on the level or priority of parameters in COVID-19 susceptibility analysis and mapping pandemic accurately, support decisions, and preventive actions. This study attempts to conduct COVID-19 susceptibility analysis and mapping of the COVID-19 pandemic based on socio-economic factors in the Development Unit II Kertajaya and III Tambak Wedi Surabaya City. This study uses statistical data on the number of cases of COVID-19 sufferers from March 2020 to August 2021 and data on socio-economic facilities in the form of the number of health facilities (hospitals, health centers, and clinics), tourist attractions (tourist attractions and city parks), banking (atm and bank), market, mall, gas station, and terminal. The results of random forest processing obtained the level of influence of each parameter on the spread of COVID-19, namely health facilities 24.135%, terminals 20.338%, tourist attractions 19.916%, malls 19.574%, markets 11.317%, banking 2.628%, and gas stations 2.092%. Based on previous research, the accuracy of the resulting model can be said to be good with an accuracy value of 0.946, a kappa value of 0.892, and an AUC value of 0.984. The mapping results from the random forest model, obtained areas with a low risk level in the northern part of Kenjeran Sub-District and the eastern part of Sukolilo Sub-District and Mulyorejo Sub-District, while the high risk level is in the middle part of Mulyorejo Sub-District and the middle part of Sukolilo Sub-District. Based on these results, it is hoped that the government can maximize the social restriction program with a focus on regional concentration according to the level of susceptibility so that the control of the COVID-19 pandemic can be on target. In addition, the public must also participate in supporting programs implemented by the government by maintaining contact or social interaction, especially at socio-economic facilities which are the source of the spread of COVID-19 which is quite risky on the spread of the COVID-19 pandemic.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSG 025.069 1 Der a-1 2022 |
| Uncontrolled Keywords: | COVID-19, Random forest, Sosial-ekonomi. COVID-19, Random Forest, Sosio-Economi. |
| Divisions: | Faculty of Civil, Planning, and Geo Engineering (CIVPLAN) > Geomatics Engineering > 29202-(S1) Undergraduate Thesis |
| Depositing User: | Mr. Marsudiyana - |
| Date Deposited: | 20 May 2026 01:57 |
| Last Modified: | 20 May 2026 01:57 |
| URI: | http://repository.its.ac.id/id/eprint/133252 |
Actions (login required)
 |
View Item |