Implementasi Metode Random Forest Untuk Klasifikasi Pasien Yang Terkena Gagal Jantung

Nabila, Zulfika Hemadewi (2023) Implementasi Metode Random Forest Untuk Klasifikasi Pasien Yang Terkena Gagal Jantung. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text COVER-5-15_merged.pdf Download (372kB)
	Text ABSTRACT (1).pdf Download (11kB)
	Text ABSTRAK (1).pdf Download (10kB)
	Text BAB I (1).pdf Download (95kB)
	Text BAB II (1).pdf Download (357kB)
	Text BAB III (1).pdf Download (392kB)
	Text BAB IV (1).pdf Restricted to Registered users only Download (512kB) | Request a copy
	Text BAB V (1).pdf Download (10kB)
	Text DAFTAR PUSTAKA.pdf Download (156kB)
	Text LAMPIRAN (1).pdf Restricted to Registered users only Download (1MB) | Request a copy

Abstract

The heart is one of the organs in the body that has the obligation to pump and deliver blood that transports oxygen or O2 and carbon dioxide or CO2 to all organs in the body. Heart failure is usually caused by cardiovascular disease (CVD). In Indonesia, the death rate from heart failure is very high. Based on doctors' analysis, the percentage of patients affected by heart failure in Indonesia in 2013 was 0.13% or around 229,696 people. Therefore, the latest efforts are needed to early detect heart failure patients so as to reduce mortality and prolong the patient's life. With the help of the latest technology, the history data of patients who experience heart failure can be processed in such a way as to form a correlation pattern between the latest data and the history data of patients who have experienced heart failure, in this case researchers use Heart Failure Prediction history data obtained from the Kaggle website as many as 299 data with 13 features. Researchers used the Random Forest algorithm combined with the Min Max Normalization to normalize distorted data and Adaptive Synthetic (ADASYN) to balance the data of patients affected by heart failure so that the accuracy, precision, recall and f1-score values of each scheme can be known. Researchers use the Random Forest model because it works well in classification and can efficiently clean up large amounts of training data. For the test scenario, researchers used the K-Fold Cross Validation technique with k=10 which means dividing the data into two parts, namely training data and testing data as much as 10 folds. The results obtained are that the 6-feature scheme and the 7-feature scheme have the highest accuracy value of 97.6% compared to other schemes with a precision value of 0.95, a recall of 10, and an f1-score of 0.97.

Actions (login required)

View Item