Prediksi Tingkat Keparahan Pasien Covid-19 di Rumah Sakit Universitas Airlangga Surabaya Menggunakan Combine Sampling Support Vector Machine

Siska, Mega Oktaviana (2022) Prediksi Tingkat Keparahan Pasien Covid-19 di Rumah Sakit Universitas Airlangga Surabaya Menggunakan Combine Sampling Support Vector Machine. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06211840000025-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Coronavirus Disease 19 (COVID-19) merupakan virus baru yang ditemukan dan diidentifikasi penyebab dari kasus pneumonia yang terjadi di Kota Wuhan. Penyebaran virus Covid-19 sangat cepat menyebar ke seluruh dunia. Untuk mengurangi tingginya angka kematian, observasi kondisi harian pada pasien Covid-19 menjadi hal yang penting bagi rumah sakit guna mengidentifikasi pasien yang mungkin memerlukan tindakan khusus secara cepat dan tepat. Seluruh variabel akan diprediksi menghasilkan tiga tingkat keparahan pasien Covid 19 yaitu meninggal, risiko berat (masuk ICU), dan risiko ringan. Penelitian ini bertujuan untuk mengetahui karakteristik pasien Covid-19 dan memodelkan tingkat keparahan pasien Covid-19 serta mengetahui hasil prediksi tingkat keparahan pasien Covid-19 menggunakan metode Combine Sampling SVM dengan pendekatan One-Againt-One (OAO) dan One-Againt-All (OAA). Hasil analisis pada penelitian ini menunjukkan bahwa mayoritas pasien yang meninggal usia >50 tahun keatas, berjenis kelamin laki-laki, saturasi oksigen kurang dari 90%, laju respirasi lebih dari 26 kali per menit, tekanan darah kategori 91-130 mmHg, temperature 35,1 – 37oC, dan pasien yang sebelumnya memiliki riwayat penyakit rentan terkena Covid-19 lebih parah. Sementara itu, metode Combine Sampling merupakan metode yang lebih baik dibandingkan dengan metode SMOTE dan Tomek Links dengan nilai rata-rata AUC sebesar 0,9062 pada pendekatan OAO parameter c=10 dengan prediksi tingkat keparahan pasien Covid 19 menggunakan pendekatan OAO dan OAA pada parameter c=10 yang mengindikasikan semakin besar parameter maka memperkecil terjadinya kesalahan pengklasifikasian.
==============================================================================================================================
Coronavirus Disease 19 (COVID-19) is a new virus that was discovered and identified as the cause of pneumonia cases that occurred in Wuhan City. The Covid-19 virus is spreading very quickly all over the world. To increase the mortality rate, daily condition observations in Covid-19 patients are important for hospitals to identify conditions that may require special actions quickly and appropriately. All variables will predict three levels of Covid-19 disease severity, namely death, severe risk (ICU admission), and mild risk. This study aims to determine the characteristics of Covid-19 patients and model the severity of Covid-19 patients and determine the results of predicting the severity of Covid-19 using the Combine Sampling SVM method with One-Againt-One (OAO) and One-Againt-All (OAA) approaches. The results of the analysis in this study showed that the majority of patients who died >50 years and over, were male, oxygen saturation was less than 90%, respiration rate was more than 26 times per minute, blood pressure category 91-130 mmHg, temperature 35, 1 – 37oC, and patients who previously had a history of illness were more susceptible to COVID-19. Meanwhile, the Combine Sampling method is a better method than the SMOTE and Tomek Links methods with an average AUC value of 0.9062 on the OAO parameter approach c = 10 with predictions of the severity of Covid-19 patients using the OAO and OAA approaches on the parameters c = 10 which increases the parameter size, thereby reducing the occurrence of classification errors.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Combine Sampling Support Vector Machine, Covid-19, Keparahan, Prediksi. Combine Sampling Support Vector Machine, Covid-19, Severity, Prediction.
Subjects:	H Social Sciences > HA Statistics
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	10 Jun 2026 04:16
Last Modified:	10 Jun 2026 04:16
URI:	http://repository.its.ac.id/id/eprint/133686

Actions (login required)

View Item