PENERAPAN ALGORITMA RELIEFF DAN BACKPROPAGATION NEURAL NETWORK (BNN) DALAM PENANGANAN DATA HILANG DAN KLASIFIKASI PENYAKIT JANTUNG

SYAFITRI, HENNY NUR and Desiani, Anita and Irmeilyana, Irmeilyana (2022) PENERAPAN ALGORITMA RELIEFF DAN BACKPROPAGATION NEURAL NETWORK (BNN) DALAM PENANGANAN DATA HILANG DAN KLASIFIKASI PENYAKIT JANTUNG. Undergraduate thesis, Sriwijaya University.

	Text RAMA_44201_08011281823045.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_44201_08011281823045_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (8MB) | Request a copy
Preview	Text RAMA_44201_08011281823045_0011127702_0017057403_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (730kB) | Preview
	Text RAMA_44201_08011281823045_0011127702_0017057403_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (587kB) | Request a copy
	Text RAMA_44201_08011281823045_0011127702_0017057403_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (338kB) | Request a copy
	Text RAMA_44201_08011281823045_0011127702_0017057403_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (538kB) | Request a copy
	Text RAMA_44201_08011281823045_0011127702_0017057403_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (421kB) | Request a copy
	Text RAMA_44201_08011281823045_0011127702_0017057403_06_ref.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (276kB) | Request a copy

Abstract

Classification is the process of classifying an object based on a certain model. This process is often used in various fields, such as health, environment, and data research. One application of classification in the health sector is to predict factors for heart disease. The dataset used to predict cardiovascular disease factors comes from the University of California Irvine (UCI). This dataset has a weakness there is missing data in some features. Missing data can be overcome by applying the ReliefF algorithm, where each feature is sorted from the largest weight to the smallest weight. The advantage of the ReliefF algorithm is that it does not limit the data types used and can effectively deal with multiclass problems, missing data, and noise tolerance. In this research, the application of the ReliefF algorithm and testing using the Backpropagation Neural Network (BNN) method was carried out to determine the best features of the predictive factors for heart disease. Before applying the ReliefF algorithm in the detection of heart disease, the accuracy, precision, and recall values obtained were 77.21%, 72.31%, and 76.60%. Meanwhile, after applying the ReliefF algorithm, the performance evaluation results obtained were 84.01%, 80.09%, and 84.76%. Based on the performance results obtained through testing using the Backpropagation Neural Network (BNN) method, these values indicate that the ReliefF algorithm can be used to select the most influential and good features to increase the level of accuracy in predicting heart disease patients.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Klasifikasi, penyakit jantung, data hilang, algoritma ReliefF, Backpropagation Neural Network (BNN)
Subjects:	Q Science > QA Mathematics > QA273-280 Probabilities. Mathematical statistics
Divisions:	08-Faculty of Mathematics and Natural Science > 44201-Mathematics (S1)
Depositing User:	Henny Nur Syafitri
Date Deposited:	01 Jul 2022 02:42
Last Modified:	01 Jul 2022 02:42
URI:	http://repository.unsri.ac.id/id/eprint/73145

Actions (login required)

View Item