Open AccessDeteksi Kanker Berdasarkan Klasifikasi Microarray Data
Author(s) -
Adiwijaya Adiwijaya
Publication year - 2018
Publication title -
jurnal media informatika budidarma/jurnal media informatika budidarma
Language(s) - English
Resource type - Journals
eISSN - 2614-5278
pISSN - 2548-8368
DOI - 10.30865/mib.v2i4.1043
Subject(s) - microarray analysis techniques , dimensionality reduction , dimension (graph theory) , artificial intelligence , dna microarray , computer science , artificial neural network , microarray , pattern recognition (psychology) , data mining , gene , gene expression , biology , mathematics , genetics , pure mathematics
Cancer is one of the diseases that can cause human death in the world and become the biggest cause of death after heart disease. Therefore we need a DNA microarray technology which is used to examine how gene expression patterns change under different conditions, so that the technology is able to detect a person with cancer or not with accurate analysis. The size of the dimension in the microarray data can affect the gene expression analysis that is used to find informative genes, for that we need a good method of dimension reduction and classification so that it can get the best results and accuracy. Many techniques can be applied in DNA microarray, one of them is BPNN Back Propagation Neural Network as a classification and PCA as dimension reduction, where both have been tested in several previous studies. By applying BPNN and PCA on several types of cancer data, it was found that BPNN and PCA get more than 80% accuracy results with training time 0-4 seconds.