Abstract

Alzheimer is a degenerative disease and one of the most common cases of dementia. One of the important keys to treat Alzheimer is an early detection which can be done by analyzing the genes expression in the DNA by using DNA microarray technology. The basic problem in classification is to find out the best method which is able to accurately predict the case. This reearch applies Logistic Regression Ensemble (LORENS) to classify Alzheimer related genes and compare the result with Naïve Bayes classifier. This research examines 178 observations consisting of 2 classes where 98 observations are Alzheimer's genes and 80 observations are normal genes. The analysis shows that LORENS outperforms the Naïve Bayes classifier evaluated with Cross Validation. The best LORENS setting is obtained for 5 partitions and threshold 0.5 which leads to 75.28% accuracy and 0.759 for the Area Under Curve (AUC). This results indicate that LORENS is a good appoach to classify Alzheimer gene expression.

Original languageEnglish
Title of host publicationProceeding of 2017 International Conference on Smart Cities, Automation and Intelligent Computing Systems, ICON-SONICS 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages36-41
Number of pages6
ISBN (Electronic)9781509062805
DOIs
Publication statusPublished - 2 Jul 2017
Event2017 International Conference on Smart Cities, Automation and Intelligent Computing Systems, ICON-SONICS 2017 - Yogyakarta, Indonesia
Duration: 8 Nov 201710 Nov 2017

Publication series

NameProceeding of 2017 International Conference on Smart Cities, Automation and Intelligent Computing Systems, ICON-SONICS 2017
Volume2018-January

Conference

Conference2017 International Conference on Smart Cities, Automation and Intelligent Computing Systems, ICON-SONICS 2017
Country/TerritoryIndonesia
CityYogyakarta
Period8/11/1710/11/17

Keywords

  • Alzheimer
  • Bayes
  • LORENS
  • accuracy
  • training

Fingerprint

Dive into the research topics of 'Logistic regression ensemble to classify Alzheimer gene expression'. Together they form a unique fingerprint.

Cite this