Optimasi Metode Support Vector Machine Linear Dengan Algoritma Genetika Pada Klasifikasi Tingkat Obesitas

Azizah, Ratih Nuur (2024) Optimasi Metode Support Vector Machine Linear Dengan Algoritma Genetika Pada Klasifikasi Tingkat Obesitas. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 20081010136.-cover.pdf Download (1MB)
	Text (Bab 1) 20081010136.-bab1.pdf Download (164kB)
	Text (Bab 2) 20081010136.-bab2.pdf Restricted to Repository staff only until 3 June 2028. Download (243kB)
	Text (Bab 3) 20081010136.-bab3.pdf Restricted to Repository staff only until 3 June 2028. Download (482kB)
	Text (Bab 4) 20081010136.-bab4.pdf Restricted to Repository staff only until 3 June 2028. Download (1MB)
	Text (Bab 5) 20081010136.-bab5.pdf Download (47kB)
	Text (Daftar Pustaka) 20081010136.-daftarpustaka.pdf Download (118kB)

Abstract

Obesity has become a serious problem around the world. Obesity can also trigger various other diseases such as diabetes, heart disease and cancer. Irregular diet and physical activity is one of the factors of obesity. Therefore, understanding the relationship between diet, physical condition, and obesity levels is crucial for the development of effective obesity prevention strategies. In this journal, classification is done using Linear Support Vector Machine method along with Genetic Algorithm optimization method is chosen to help tackle the problem. The data used is 2111 with 17 variables. Meanwhile, there are 7 classes in this dataset which include Insufficient Weight, Normal Weight, Obesity Type I, Obesity Type II, Obesity Type III, Over Weight I and Over Weight II. In the division of training data and test data, there are three tests carried out, namely 70:30, 80:20, and 90:10. Then in Genetic Algorithm optimization, population variations are used at 5, 10, and 25. In accordance with the results of the research, the highest accuracy was obtained in the Linear Support Vector Machine method with Genetic Algorithm optimization of 97.9% with a population size of 10 and the division of test and training data of 80:20. Based on these results, the Linear Support Vector Machine method optimized with the Genetic Algorithm is able to classify the level of obesity and can increase its accuracy value.

Item Type:

Thesis (Undergraduate)

Contributors: