Analisis Perbandingan Klasifikasi Support Vector Machine (SVM) dan K-Nearest Neighbors (KNN) untuk Deteksi Kanker dengan Data Microarray

Abstract

Cancer is a disease that can cause human death in various countries. According to WHO in 2018, cancer causes 9.6 million human deaths worldwide. Globally, about 1 in 6 deaths is due to cancer. Therefore, we need a technology that can be used for cancer detection with high acuration so that cancer can be detected early. Microarrays technique can predict certain tissues in humans and can be classified as cancer or not. However, microarray data has a problem with very large dimensions. To overcome this problem, in this study use one of the dimension reduction techniques, namely Partial Least Square(PLS) and use Support vector Machine (SVM) and K-Nearest Neighbors as a classification method, which will be used to compare which is better.The system built was able to reach 98.54% in leukemia data with PLS-KNN, 100% in lung data with KNN, 66.52% in breast data with PLS-KNN, and 85.60% in colon data with PLS- SVM. KNN is able to get the best in three data from four valued data.

Perbandingan Klasifikasi Data Calon Pendonor Darah dengan Algoritma K-Nearest Neighbor dan Support Vector Machine

“…Ada beberapa parameter-parameter yang akan digunakan antara lainnya seperti C= [1,10,100] , degree= [2,3,4], dan gamma=[1, 0.1, 0.01]. Pada penelitian ini komposisi perbandingan data traning dan data testing dengan menggunakan algoritma Knn dan SvM akan dilakukan dengan tiga kali percobaan perhitungan yaitu dengan menentukan nilai data testing sebesar =[0.2 , 0.3 , dan 0.4] [14]. Perbandingan tersebut adalah 80% data training dan 20% data testing, serta 70% data training dan 30% data testing, yang terakhir yaitu 60% data training dan 40% data testing.…”

Algoritma K-Nearest Neighbor dengan Euclidean Distance dan Manhattan Distance untuk Klasifikasi Transportasi Bus

“…Selanjutnya pembagian dataset data training dan data data testing. Setelah pembagian data menjadi dua bagian training dan testing dilakukan langkah seleksi Class dengan menggunakan Algoritma K-NN dengan dua metode pendekatan Euclidean Distance dan Manhattan Distance [4].…”

Comparative analysis of support vector machine and k-nearest neighbors with a pyramidal histogram of the gradient for sign language detection

“…The SVM and KNN classifications have been studied by [8] using Partial Least Square (PLS) as dimension reduction, SVM, and KNN as classifications. This study produced the highest accuracy of 98.54% on leukemia data with PLS-KNN, 100% on lung data with KNN, 66.52% on breast data with PLS-KNN, and 85.60% on colon data with PLS-SVM.…”