A one class classifier for Signal identification: a biological case study

Risultato della ricerca: Other

6 Citazioni (Scopus)

Abstract

The paper describes an application of a one-class KNN to identify different signal patterns embedded in a noisestructured background. The problem become harder whenever only one pattern is well represented in the signal, insuch cases one class classifier techniques are more indicated. The classification phase is applied after apreprocessing phase based on a Multi Layer Model (MLM) that provides a preliminary signal segmentation in aninterval feature space. The one-class KNN has been tested on synthetic data that simulate microarray data for theidentification of nucleosomes and linker regions across DNA. Results have shown a good recognitionrate on synthetic data for nucleosome and linker regions.
Lingua originaleEnglish
Pagine747-754
Numero di pagine6
Stato di pubblicazionePublished - 2008

Fingerprint

Microarrays
DNA
Classifiers
Classifier
Synthetic Data
Microarray Data
Feature Space
Multilayer
Segmentation
Class
Model

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cita questo

@conference{1c76a986236643beba0c56c0ca3234b3,
title = "A one class classifier for Signal identification: a biological case study",
abstract = "The paper describes an application of a one-class KNN to identify different signal patterns embedded in a noisestructured background. The problem become harder whenever only one pattern is well represented in the signal, insuch cases one class classifier techniques are more indicated. The classification phase is applied after apreprocessing phase based on a Multi Layer Model (MLM) that provides a preliminary signal segmentation in aninterval feature space. The one-class KNN has been tested on synthetic data that simulate microarray data for theidentification of nucleosomes and linker regions across DNA. Results have shown a good recognitionrate on synthetic data for nucleosome and linker regions.",
keywords = "Bioinformatics, Multi Layer Method, Nucleosome Positioning., One Class classification",
author = "{Di Gesu'}, Vito and {Lo Bosco}, Giosue' and Luca Pinello",
year = "2008",
language = "English",
pages = "747--754",

}

TY - CONF

T1 - A one class classifier for Signal identification: a biological case study

AU - Di Gesu', Vito

AU - Lo Bosco, Giosue'

AU - Pinello, Luca

PY - 2008

Y1 - 2008

N2 - The paper describes an application of a one-class KNN to identify different signal patterns embedded in a noisestructured background. The problem become harder whenever only one pattern is well represented in the signal, insuch cases one class classifier techniques are more indicated. The classification phase is applied after apreprocessing phase based on a Multi Layer Model (MLM) that provides a preliminary signal segmentation in aninterval feature space. The one-class KNN has been tested on synthetic data that simulate microarray data for theidentification of nucleosomes and linker regions across DNA. Results have shown a good recognitionrate on synthetic data for nucleosome and linker regions.

AB - The paper describes an application of a one-class KNN to identify different signal patterns embedded in a noisestructured background. The problem become harder whenever only one pattern is well represented in the signal, insuch cases one class classifier techniques are more indicated. The classification phase is applied after apreprocessing phase based on a Multi Layer Model (MLM) that provides a preliminary signal segmentation in aninterval feature space. The one-class KNN has been tested on synthetic data that simulate microarray data for theidentification of nucleosomes and linker regions across DNA. Results have shown a good recognitionrate on synthetic data for nucleosome and linker regions.

KW - Bioinformatics

KW - Multi Layer Method

KW - Nucleosome Positioning.

KW - One Class classification

UR - http://hdl.handle.net/10447/40102

M3 - Other

SP - 747

EP - 754

ER -