Learning the relative importance of features in image data

Aparna Varde, Elke Rundensteiner, Giti Javidi, Ehsan Sheybani, Jianyu Liang

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

3 Citations (Scopus)

Abstract

In computational analysis in scientific domains, images are often compared based on their features, e.g., size, depth and other domain-specific aspects. Certain features may be more significant than others while comparing the images and drawing corresponding inferences for specific applications. Though domain experts may have subjective notions of similarity for comparison, they seldom have a distance function that ranks the image features based on their relative importance. We propose a method called FeaturesRank for learning such a distance function in order to capture the semantics of the images. We are given training samples with pairs of images and the extent of similarity identified for each pair. Using a guessed initial distance function, FeaturesRank clusters the given images in levels. It then adjusts the distance function based on the eiror between the clusters and training samples using heuristics proposed in this paper. The distance function that gives the lowest error is the output. This contains the features ranked in the order most appropriate the domain. FeaturesRank is evaluated with real image data from nanotechnology and bioinformatics. The results of our evaluation are presented in the paper.

Original languageEnglish
Title of host publicationWorkshops in Conjunction with the International Conference on Data Engineering - ICDE' 07
Pages237-244
Number of pages8
DOIs
StatePublished - 1 Dec 2007
EventWorkshops in Conjunction with the 23rd International Conference on Data Engineering - ICDE 2007 - Istanbul, Turkey
Duration: 15 Apr 200720 Apr 2007

Other

OtherWorkshops in Conjunction with the 23rd International Conference on Data Engineering - ICDE 2007
CountryTurkey
CityIstanbul
Period15/04/0720/04/07

Fingerprint

Bioinformatics
Nanotechnology
Semantics

Cite this

Varde, A., Rundensteiner, E., Javidi, G., Sheybani, E., & Liang, J. (2007). Learning the relative importance of features in image data. In Workshops in Conjunction with the International Conference on Data Engineering - ICDE' 07 (pp. 237-244). [4400998] https://doi.org/10.1109/ICDEW.2007.4400998
Varde, Aparna ; Rundensteiner, Elke ; Javidi, Giti ; Sheybani, Ehsan ; Liang, Jianyu. / Learning the relative importance of features in image data. Workshops in Conjunction with the International Conference on Data Engineering - ICDE' 07. 2007. pp. 237-244
@inproceedings{0c4254f3b38248a68abb147d8efb3c5d,
title = "Learning the relative importance of features in image data",
abstract = "In computational analysis in scientific domains, images are often compared based on their features, e.g., size, depth and other domain-specific aspects. Certain features may be more significant than others while comparing the images and drawing corresponding inferences for specific applications. Though domain experts may have subjective notions of similarity for comparison, they seldom have a distance function that ranks the image features based on their relative importance. We propose a method called FeaturesRank for learning such a distance function in order to capture the semantics of the images. We are given training samples with pairs of images and the extent of similarity identified for each pair. Using a guessed initial distance function, FeaturesRank clusters the given images in levels. It then adjusts the distance function based on the eiror between the clusters and training samples using heuristics proposed in this paper. The distance function that gives the lowest error is the output. This contains the features ranked in the order most appropriate the domain. FeaturesRank is evaluated with real image data from nanotechnology and bioinformatics. The results of our evaluation are presented in the paper.",
author = "Aparna Varde and Elke Rundensteiner and Giti Javidi and Ehsan Sheybani and Jianyu Liang",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/ICDEW.2007.4400998",
language = "English",
isbn = "1424408326",
pages = "237--244",
booktitle = "Workshops in Conjunction with the International Conference on Data Engineering - ICDE' 07",

}

Varde, A, Rundensteiner, E, Javidi, G, Sheybani, E & Liang, J 2007, Learning the relative importance of features in image data. in Workshops in Conjunction with the International Conference on Data Engineering - ICDE' 07., 4400998, pp. 237-244, Workshops in Conjunction with the 23rd International Conference on Data Engineering - ICDE 2007, Istanbul, Turkey, 15/04/07. https://doi.org/10.1109/ICDEW.2007.4400998

Learning the relative importance of features in image data. / Varde, Aparna; Rundensteiner, Elke; Javidi, Giti; Sheybani, Ehsan; Liang, Jianyu.

Workshops in Conjunction with the International Conference on Data Engineering - ICDE' 07. 2007. p. 237-244 4400998.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

TY - GEN

T1 - Learning the relative importance of features in image data

AU - Varde, Aparna

AU - Rundensteiner, Elke

AU - Javidi, Giti

AU - Sheybani, Ehsan

AU - Liang, Jianyu

PY - 2007/12/1

Y1 - 2007/12/1

N2 - In computational analysis in scientific domains, images are often compared based on their features, e.g., size, depth and other domain-specific aspects. Certain features may be more significant than others while comparing the images and drawing corresponding inferences for specific applications. Though domain experts may have subjective notions of similarity for comparison, they seldom have a distance function that ranks the image features based on their relative importance. We propose a method called FeaturesRank for learning such a distance function in order to capture the semantics of the images. We are given training samples with pairs of images and the extent of similarity identified for each pair. Using a guessed initial distance function, FeaturesRank clusters the given images in levels. It then adjusts the distance function based on the eiror between the clusters and training samples using heuristics proposed in this paper. The distance function that gives the lowest error is the output. This contains the features ranked in the order most appropriate the domain. FeaturesRank is evaluated with real image data from nanotechnology and bioinformatics. The results of our evaluation are presented in the paper.

AB - In computational analysis in scientific domains, images are often compared based on their features, e.g., size, depth and other domain-specific aspects. Certain features may be more significant than others while comparing the images and drawing corresponding inferences for specific applications. Though domain experts may have subjective notions of similarity for comparison, they seldom have a distance function that ranks the image features based on their relative importance. We propose a method called FeaturesRank for learning such a distance function in order to capture the semantics of the images. We are given training samples with pairs of images and the extent of similarity identified for each pair. Using a guessed initial distance function, FeaturesRank clusters the given images in levels. It then adjusts the distance function based on the eiror between the clusters and training samples using heuristics proposed in this paper. The distance function that gives the lowest error is the output. This contains the features ranked in the order most appropriate the domain. FeaturesRank is evaluated with real image data from nanotechnology and bioinformatics. The results of our evaluation are presented in the paper.

UR - http://www.scopus.com/inward/record.url?scp=48349093471&partnerID=8YFLogxK

U2 - 10.1109/ICDEW.2007.4400998

DO - 10.1109/ICDEW.2007.4400998

M3 - Conference contribution

SN - 1424408326

SN - 9781424408320

SP - 237

EP - 244

BT - Workshops in Conjunction with the International Conference on Data Engineering - ICDE' 07

ER -

Varde A, Rundensteiner E, Javidi G, Sheybani E, Liang J. Learning the relative importance of features in image data. In Workshops in Conjunction with the International Conference on Data Engineering - ICDE' 07. 2007. p. 237-244. 4400998 https://doi.org/10.1109/ICDEW.2007.4400998