Citation

Mohd Ali, Nursabillilah and Besar, Rosli and Ab Aziz, Nor Azlina (2021) Microarray Breast Cancer Classification using Machine Learning Method. In: 2nd FET PG Engineering Colloquium Proceedings 2021, 1-15 Dec. 2021, Online Conference. (Unpublished)

Text 09 Nursabillilah.pdf Restricted to Repository staff only Download (17kB)

Abstract

Breast cancer is one of the leading causes of death and most frequently diagnosed cancer amongst women. It affects patients that live in average living condition that must go beyond expensive and distressing medications. The disease involves one in eight women worldwide. Annually, almost half a million women do not survive the disease and die from the cancer. Machine learning is data analysis methods that is efficiently demonstrated good solutions and outperforms other results in predicting breast cancer using built on previously trained data. In this work, multi-machine learning method is applied for microarray gene expression breast cancer classification. The data is processed prior to classify using machine learning method. The machine learning methods utilised are support vector machine (SVM), random forest (RF), decision tree (DT), logistic regression (LR), k-nearest neighbor (KNN) and naïve bayes (NB). The classification rate is compared among the classifiers applied. Experimental result shows some of microarray breast cancer classification able to demonstrate good accuracy rate. However, the classification performance rate according to the dataset size and ratio of classes. For instance, GSE1456 using DT classifier before tuning the classifier with hyperparameter showed highest rate of 84.38% classification accuracy. Whereas when using other models, GSE1456 performed classification rate approximately at 75%~78%. SVMs and RFs models generated the best classification accuracy in the overall datasets. The microarray breast cancer comes with large number of features with small sample size, the dataset dimension should be reduced to gain better classification rate prior to classification using machine learning method. This work evaluates multi-classifiers for breast cancer classification using microarray data. The finding is useful in development of microarray-based breast cancer classification system.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Microarray dataset, breast cancer, classification, machine learning
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions:	Faculty of Engineering and Technology (FET)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	25 Jan 2022 14:09
Last Modified:	25 Jan 2022 14:09
URII:	http://shdl.mmu.edu.my/id/eprint/9880

Downloads

Edit (login required)