KLASIFIKASI DATA MICROARRAY MENGGUNAKAN DISCRETE WAVELET TRANSFORM DAN EXTREME LEARNING MACHINE
Cancer can be classified based on its morphologis structure or gene expression values in microarray data. Cancer classification based on its morphologis structure is difficult because the poor distinction of morphologis structures among different classes of cancer, so in this research cancer is clas...
| Main Authors: | , |
|---|---|
| Format: | Thesis |
| Published: |
[Yogyakarta] : Universitas Gadjah Mada
2014
|
| Subjects: |
| _version_ | 1826048159741116416 |
|---|---|
| author | , KHADIJAH , Dra. Sri Hartati, M.Sc., Ph.D. |
| author_facet | , KHADIJAH , Dra. Sri Hartati, M.Sc., Ph.D. |
| author_sort | , KHADIJAH |
| collection | UGM |
| description | Cancer can be classified based on its morphologis structure or gene
expression values in microarray data. Cancer classification based on its
morphologis structure is difficult because the poor distinction of morphologis
structures among different classes of cancer, so in this research cancer is
classified based on the gene expression value in microarray data. The important
things in the microarray data classification are the huge number of genes as
dimension of microarray data (high dimensionality) and the limited sample size,
so the method of dimension reduction and the classifier algorithm should be well
determined.
The aim of this research is to bulid microarray data classifier. The
classification process is started by reducing dimension of microarray data using
DWT. It can be done by decomposing the samples until certain decomposition
level and then use the approximation coefficients at those level as features set to
the classifier. Classifier used in this research is ELM implemeted on RBFN.
Dataset used in this research are GCM (16.063 genes, 14 classes) and Subtypes-
Leukemia (12.600 genes, 7 classes).
Testing process is done by randomly dividing the training and testing data
ten times with same proprotion of training and testing data. The result achieved by
classifier for GCM dataset is not quite good, mean of accuracy 75,65% ± 6,86%
and minimum sensitivity 20% ± 17,21%. The low value of minimum sensitivity
indicate that the sensitivity among all classes is not well averaged. The result
achieved by classifier for Subtypes-Leukemia dataset with smaller number of
classes, is quite better than GCM, mean of accuracy 89,29% ± 2,42% and mean
of minimum sensitivity 54,44% ± 25,5%. The low value of minimum sensitivity
resulted from BCR-ABL class because of the smallest sample number of that class
among the other classes. |
| first_indexed | 2024-03-13T23:23:36Z |
| format | Thesis |
| id | oai:generic.eprints.org:129109 |
| institution | Universiti Gadjah Mada |
| last_indexed | 2024-03-13T23:23:36Z |
| publishDate | 2014 |
| publisher | [Yogyakarta] : Universitas Gadjah Mada |
| record_format | dspace |
| spelling | oai:generic.eprints.org:1291092016-03-04T07:57:57Z https://repository.ugm.ac.id/129109/ KLASIFIKASI DATA MICROARRAY MENGGUNAKAN DISCRETE WAVELET TRANSFORM DAN EXTREME LEARNING MACHINE , KHADIJAH , Dra. Sri Hartati, M.Sc., Ph.D. ETD Cancer can be classified based on its morphologis structure or gene expression values in microarray data. Cancer classification based on its morphologis structure is difficult because the poor distinction of morphologis structures among different classes of cancer, so in this research cancer is classified based on the gene expression value in microarray data. The important things in the microarray data classification are the huge number of genes as dimension of microarray data (high dimensionality) and the limited sample size, so the method of dimension reduction and the classifier algorithm should be well determined. The aim of this research is to bulid microarray data classifier. The classification process is started by reducing dimension of microarray data using DWT. It can be done by decomposing the samples until certain decomposition level and then use the approximation coefficients at those level as features set to the classifier. Classifier used in this research is ELM implemeted on RBFN. Dataset used in this research are GCM (16.063 genes, 14 classes) and Subtypes- Leukemia (12.600 genes, 7 classes). Testing process is done by randomly dividing the training and testing data ten times with same proprotion of training and testing data. The result achieved by classifier for GCM dataset is not quite good, mean of accuracy 75,65% ± 6,86% and minimum sensitivity 20% ± 17,21%. The low value of minimum sensitivity indicate that the sensitivity among all classes is not well averaged. The result achieved by classifier for Subtypes-Leukemia dataset with smaller number of classes, is quite better than GCM, mean of accuracy 89,29% ± 2,42% and mean of minimum sensitivity 54,44% ± 25,5%. The low value of minimum sensitivity resulted from BCR-ABL class because of the smallest sample number of that class among the other classes. [Yogyakarta] : Universitas Gadjah Mada 2014 Thesis NonPeerReviewed , KHADIJAH and , Dra. Sri Hartati, M.Sc., Ph.D. (2014) KLASIFIKASI DATA MICROARRAY MENGGUNAKAN DISCRETE WAVELET TRANSFORM DAN EXTREME LEARNING MACHINE. UNSPECIFIED thesis, UNSPECIFIED. http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=69487 |
| spellingShingle | ETD , KHADIJAH , Dra. Sri Hartati, M.Sc., Ph.D. KLASIFIKASI DATA MICROARRAY MENGGUNAKAN DISCRETE WAVELET TRANSFORM DAN EXTREME LEARNING MACHINE |
| title | KLASIFIKASI DATA MICROARRAY MENGGUNAKAN DISCRETE WAVELET TRANSFORM DAN
EXTREME LEARNING MACHINE |
| title_full | KLASIFIKASI DATA MICROARRAY MENGGUNAKAN DISCRETE WAVELET TRANSFORM DAN
EXTREME LEARNING MACHINE |
| title_fullStr | KLASIFIKASI DATA MICROARRAY MENGGUNAKAN DISCRETE WAVELET TRANSFORM DAN
EXTREME LEARNING MACHINE |
| title_full_unstemmed | KLASIFIKASI DATA MICROARRAY MENGGUNAKAN DISCRETE WAVELET TRANSFORM DAN
EXTREME LEARNING MACHINE |
| title_short | KLASIFIKASI DATA MICROARRAY MENGGUNAKAN DISCRETE WAVELET TRANSFORM DAN
EXTREME LEARNING MACHINE |
| title_sort | klasifikasi data microarray menggunakan discrete wavelet transform dan extreme learning machine |
| topic | ETD |
| work_keys_str_mv | AT khadijah klasifikasidatamicroarraymenggunakandiscretewavelettransformdanextremelearningmachine AT drasrihartatimscphd klasifikasidatamicroarraymenggunakandiscretewavelettransformdanextremelearningmachine |