Higher-order statistics and neural network based multi-classifier system for gene identification
This paper presents the use of higher order statistics and a neural network based multi-classifier system for gene and exon identification of a DNA sequence. Newly proposed higher order statistics features, combined with frequency domain analysis, are used to train three different neural networks. C...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English English |
| Published: |
2007
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/2341/1/GunawanICSPCS2007final.pdf http://irep.iium.edu.my/2341/4/Program_Final.pdf http://irep.iium.edu.my/2341/ http://www.dspcs-witsp.com/icspcs_2007/cfp.html |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Universiti Islam Antarabangsa Malaysia |
| Language: | English English |
| Summary: | This paper presents the use of higher order statistics and a neural network based multi-classifier system for gene and exon identification of a DNA sequence. Newly proposed higher order statistics features, combined with frequency domain analysis, are used to train three different neural networks. Classification results of the three individual neural networks are combined through an aggregation function, of which five variants are compared herein. An evaluation of the multi-classifier system on 117 sequences from the HMR195 database shows that when different opinions of more classifiers on the same input data are integrated within a multi-classifier system, a relative improvement in precision of 5% over the individual performances of the neural networks can be obtained. |
|---|