Friday 29th of March 2024
 

An Efficient Diseases Classifier based on Microarray Datasets using Clustering ANOVA Extreme Learning Machine (CAELM)


Shamsan Aljamali, Zhang Zuping and Long Jun

Cancer is a group of diseases distinguished by unregulated growth and spread of cells which has become one of the most dangerous diseases. As a result of the victims of cancer are increasing steadily, the necessity is increasing to find classification techniques for cancer diseases. The present study is aimed to obtain better results of the classification model with high accuracy. Herein, we proposed a method of developing an efficient classifier based on microarray datasets. Moreover, we focused on accuracy, dimensionality reduction and fast classification issues. The proposed method Clustering ANOVA Extreme Learning Machine (CAELM) is a hybrid approach based on Extreme Leaning Machine with RBF kernel function. This hybrid approach consist of two phases: data preprocessing (normalization and genes selection) and data classifying. K-mean clustering was utilized as a method for clustering microarray datasets into three groups, then ANOVA were applied to analysis of variance between this groups to pick out the significant genes which were used in classification process. In case combining clustering with statistical analysis (CAELM) a much better classification accuracy is given of 95,94,100% for leukemia ,prostate and ovarian respectively . In addition, the proposed approach reduced time complexity with good performance.

Keywords: CAELM, RBF kernel, K-mean, ANOVA, microarray, genes selection, cancer classification.

Download Full-Text


ABOUT THE AUTHORS

Shamsan Aljamali
School of Information Science and Engineering, Central South University, Changsha, 410083, China

Zhang Zuping
School of Information Science and Engineering, Central South University, Changsha, 410083, China

Long Jun
School of Information Science and Engineering, Central South University, Changsha, 410083, China


IJCSI Published Papers Indexed By:

 

 

 

 
+++
About IJCSI

IJCSI is a refereed open access international journal for scientific papers dealing in all areas of computer science research...

Learn more »
Join Us
FAQs

Read the most frequently asked questions about IJCSI.

Frequently Asked Questions (FAQs) »
Get in touch

Phone: +230 911 5482
Email: info@ijcsi.org

More contact details »