PENERAPAN METODE SUPPORT VECTOR MACHINE (SVM) UNTUK KLASIFIKASI DATA EKSPRESI GEN MICROARRAY

Abstract

Pancreatic cancer is the most common malignancy of the pancreas, where the cells in the pancreas grow
and multiply abnormally and the condition spreads rapidly to nearby organs and is difficult to detect
because it does not cause immediate symptoms. Pancreatic cancer is currently the fourth leading cause of
death in developed countries. As technology advances, a new field of science has now developed, namely
bioinformatics. One of the implementations of bioinformatics is the use of computational, mathematical
and statistical methods to help solve biological problems through the analysis of gene expression data. A
computational technique for large-scale bioinformatics data to classify different types of gene expression
samples from microarray data. Microarray technology is useful for identifying cancer genes and finding
new biomarkers. In this study, classification analysis was carried out on microarray data resulting from
the expression gene of tumor tissue and normal tissue in pancreatic cancer patients using the Support
Vector Machine (SVM) method for analyzing data. Based on the analysis of the method, the results show
that the best function classification with SVM method is kernel linier function with accuracy, recall,
specificity, precision, AUC and error values of 100%, 100%, 100%, 100%, 1 and 0%.
Keywords : Pancreatic Cancer, Bioinformatics, Microarray, Classification, SVM