Amrodh Alawy, Gholiqul (2021) "PEMODELAN JUMLAH KASUS COVID-19 DENGAN MOBILITAS ORANG DAN KENDARAAN DI KOTA SURABAYA ". Magister thesis, Universitas Brawijaya.
Abstract
Mobilitas masyarakat Surabaya pada tahun 2020 mengalami fluktuasi. Pola aktivitas dan mobilitas masyarakat dibedakan menjadi 5 fase, yaitu kondisi normal (F0), fase awal pandemi (F1), fase PSBB (F2), fase transisi AKB (F3), dan fase AKB (F4). Analisis data pada penelitian ini dimulai dari analisis korelasi antara variabel jumlah kasus COVID-19 dengan mobilitas orang/kendaraan dengan Pearson Correlation (PCC) yang bertujuan untuk mengetahui sejauh apa hubungan antara kedua variabel tersebut. Dari hasil korelasi diketahui bahwa mobilitas di Terminal Purabaya dan Stasiun Gubeng memiliki korelasi sangat tinggi dengan jumlah kasu, dibandingkan mobilitas di Gerbang Tol Waru. Analisis selanjutnya adalah pembuatan model antara jumlah kasus COVID-19 dengan mobilitas orang/kendaraan dengan beberapa model regresi, yaitu regresi LINEAR, LASSO (Least Absolute Shrinkage and Selection Operator) SVM (Support Vector Machine) dan KNN (K-nearest neighbors). Penggunaan beberapa model regresi difungsikan untuk mengetahui model mana yang paling baik untuk memprediksi jumlah kasus COVID-19 per hari dengan input data mobilitas orang dan kendaraan. Model terbaik yang didapatkan adalah dengan metode K-Nearest Neighbour (KNN). Model di Surabaya didapatkan parameter k = 3 dan menggunakan 6 variabel yang menghasilkan model dengan nilai R2 = 0.821, RMSE = 0.414, dan MAE = 0.292.
English Abstract
Community mobility has a close relationship with the spread of the corona virus during the COVID-19 pandemic. Various studies have been conducted to show that community mobility has a high correlation with the spread of the virus. The government's efforts to suppress the spread of the corona virus have been carried out, one of them is by issuing a policy of limiting activities outside the home and suggesting to do activities at home (work from home). The purpose of this study is to examine the relationship between community mobility and the number of COVID-19 cases during the pandemic in order to assist the government in issuing policies. The mobility of Surabaya’s people in 2020 is changing due to government policies and COVID-19. Starting from the beginning of the year, mobility was still running normally, then it decreased drastically due to the first COVID-19 case on March 17, 2020. The government issued a policy of Large-Scale Social Restrictions (PSBB), so that mobility greatly decreased. After the PSBB period, mobility slowly increases and becomes stable during the Adaptation of New Habits (AKB). The pattern of community activity and mobility is divided into 5 phases, namely normal conditions (F0), the initial phase of the pandemic (F1), the PSBB phase (F2), the AKB transition phase (F3), and the AKB phase (F4). The data analysis in this study started from the correlation analysis between the variable number of COVID-19 cases and the mobility of people/vehicles with the Pearson Correlation (PCC) which aims to determine the extent of the relationship between the two variables. The results of the correlation analysis show that in the early phase of the pandemic (F0) and the transitional phase of AKB (F3) mobility and the number of cases have a high correlation. However, during the PSBB period (F2), mobility was not highly correlated with the number of COVID-19 cases. From the correlation results, it is also known that mobility at Purabaya Terminal and Gubeng Station has a very high correlation with the number of cases, compared to mobility at the Waru Toll Gate. This shows that the spread of the corona virus through public transportation is more vulnerable than private transportation. The next analysis is to build a model between the number of COVID-19 cases and the mobility of people/vehicles with several regression models, namely LINEAR, LASSO (Least Absolute Shrinkage and Selection Operator) SVM (Support Vector Machine) and KNN (K-nearest neighbors). The use of multiple regression models is used to find out which model is the best for predicting the number of COVID-19 cases per day by inputting data on mobility of people and vehicles. The best model obtained is the K-Nearest Neighbor (KNN) method. The model in Surabaya obtained parameter k = 3 and used 6 variables which resulted in a model with a value of R2 = 0.821, RMSE = 0.414, and MAE = 0.292. While the model in East Java obtained the parameter value of k = 3 and used 5 variables which resulted in a model with a value of R2 = 0.913, RMSE = 0.290, and MAE = 0.179. The disadvantage of this model is that there is no variable coefficient output, so the relationship between variables cannot be known. Keywords: Mobility, Correlation, Regression Model, COVID-19
| Item Type: | Thesis (Magister) |
|---|---|
| Identification Number: | 624 |
| Uncontrolled Keywords: | Kata kunci: Mobilitas, Korelasi, Model Regresi, COVID-19--.Mobility, Correlation, Regression Model, COVID-19 |
| Divisions: | S2/S3 > Magister Teknik Sipil, Fakultas Teknik |
| Depositing User: | Unnamed user with username saputro |
| Date Deposited: | 20 Oct 2021 11:46 |
| Last Modified: | 24 Feb 2022 04:40 |
| URI: | http://repository.ub.ac.id/id/eprint/184394 |
![[thumbnail of DALAM MASA EMBARGO]](http://repository.ub.ac.id/style/images/fileicons/text.png) |
Text (DALAM MASA EMBARGO)
Gholiqul Amrodh Alawy.pdf Restricted to Registered users only until 31 December 2023. Download (2MB) |
Actions (login required)
 |
View Item |