Klasifikasi Tingkat Risiko Pasien COVID-19 di RS Islam Sultan Agung Semarang Menggunakan Support Vector Machine

Taibatunniswah, Naziehah (2021) Klasifikasi Tingkat Risiko Pasien COVID-19 di RS Islam Sultan Agung Semarang Menggunakan Support Vector Machine. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
06211740000090-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (2MB) | Request a copy
[img] Text
06211740000090-Undergraduate_Thesis.pdf
Restricted to Repository staff only

Download (2MB) | Request a copy

Abstract

Pada akhir tahun 2019, dunia dihebohkan oleh Coronavirus Disease (COVID-19). Angka penularan COVID-19 bertambah setiap harinya. Hingga 27 Juli 2021, kasus positif COVID-19 di Indonesia mencapai 3.239.936 orang dengan kasus kematian sebesar 86.835. Untuk mengurangi tingginya angka tersebut, observasi kondisi harian pada pasien COVID-19 menjadi hal yang sangat penting. Observasi tersebut dicatat dalam Early Warning Scoring System (EWSS) yang terdiri atas laju respirasi, saturasi oksigen, tekanan darah, laju jantung, tingkat kesadaran, dan temperatur. EWSS akan menghasilkan tingkat risiko sebagai bahan pertimbangan tindakan medis. Oleh karena itu, dilakukan penelitian mengenai tingkat risiko pasien COVID-19 dengan metode Support Vector Machine (SVM) Multiclass pendekatan One-Against-All (OAA) dan One-Against-One (OAO). Data berasal dari rekam medis RSI Sultan Agung Semarang. Tingkat risiko sebagai variabel target (Y). Variabel prediktor (X) terdiri atas 6 variabel EWSS, usia, dan jenis kelamin. Partisi data menggunakan stratified q-fold cross validation (q=5). Hasil analisis didapatkan pendekatan terbaik yakni pendekatan One-Against-One (OAO) kernel linear pada parameter c = 10 dengan nilai AUC, precision, recall, dan f1-score mencapai 100%. ===================================================================================================== At the end of 2019, the world was shocked by Corona virus Disease (COVID-19). The COVID-19 transmission rate increase every day. As of July 27, 2021, COVID-19 confirmed cases in Indonesia reached 3,239,936 cases with 86,835 deaths. To reduce this high mortality rate, daily observation of the conditions of COVID-19 patients is very important. Daily observation are recorded in the Early Warning Scoring System (EWSS) which consists of respiration rate, oxygen saturation, blood pressure, heart rate, level of conciousness, and temperature. EWSS will produce the risk level as a consideration for medical actions. Therefore, was conducted research on the risk level of COVID-19 patients using the Support Vector Machine (SVM) Multiclass method with the One-Against-All (OAA) and One-Against-One (OAO) approach. The data came from the medical record of the Sultan Agung Islamic Hospital, Semarang. The risk level as the target variable (Y). The predictor variables (X) consist of 6 variables on the EWSS, age, and gender. Data partition using stratified q-fold cross validation (q=5). The analysis result shows that the best approach was the OAO linear kernel approach at parameter c = 10 with AUC, precision, recall, and f1-score values reaching 100%.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: COVID-19, EWSS, Klasifikasi, Support Vector Machine Multiclass, Tingkat Risiko
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning.
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User: Naziehah Taibatunniswah
Date Deposited: 03 Sep 2021 02:34
Last Modified: 03 Sep 2021 02:34
URI: https://repository.its.ac.id/id/eprint/91453

Actions (login required)

View Item View Item