Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Discovery Grants Program - Individual

University of Windsor

$24,000.00

Efficient linear dimensionality reduction and optimal multi-dimensional clustering

EF 2010-11

Résumé
Pattern recognition (PR) is a field of artificial intelligence, which involves the process of taking raw data and making a decision to assign a sample to a category or class. PR systems have been extensively studied by researchers for many years, in both theoretical and application aspects - the latter includes a variety of disciplines such as medicine, molecular biology, bioinformatics, genomics, proteomics, biometrics, security and manufacturing, just to mention a few. The essence of a PR system is to devise and evaluate an efficient classifier which is able to accurately classify unknown objects. Whether or not the classes are known, the recognition process is called supervised or unsupervised. In any case, devising a classifier requires the underlying objects be represented by means of features, typically, as multi-dimensional vectors, and the underlying classifiers are derived in terms of functions or algorithms on the corresponding multi-dimensional space. The main goals of this research proposal are to devise an efficient scheme for optimal clustering, which is capable of dealing with multi-dimensional data, and develop a robust linear dimensionality reduction scheme based on statistical properties of the classes, which is capable of achieving high accurate classification for a wide range of problems, while selecting the best subset of features. Applications of the newly proposed methods will be pursued on various problems in bioinformatics, proteomics, DNA microarrays and microbial biofilms. One of these is to efficiently cluster time-series DNA microarray data using unsupervised techniques. Another problem that will be tackled using the proposed clustering methods is DNA microarray image gridding and segmentation. Protein-protein interactions will be studied in terms of finding important residues and energetic factors in interfaces, and which are correlated to classification of protein complexes and amino acids.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
261360-2009
Name nom
Rueda, Luis
Department département
Computer Science, School of
Organization id
37
Institution établissement
University of Windsor
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2010
Competition year année de concours
2009
Award amount
24000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
331
Committee name en
Computing and Information Sciences - B
Committee name fr
Informatique - B
Area of application code
802
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Information systems and technology
Area of application fr
Systèmes d'information et technologie de l'information
Research subject code
2605
Research subject group en
Robotics
Research subject group fr
Robotique
Research subject en
Pattern analysis and machine intelligence
Research subject fr
Analyse de formes et intelligence artificielle
Partie
2
Nb partie
5
Application title
Efficient linear dimensionality reduction and optimal multi-dimensional clustering
Keyword
NSERC
Application summary
Pattern recognition (PR) is a field of artificial intelligence, which involves the process of taking raw data and making a decision to assign a sample to a category or class. PR systems have been extensively studied by researchers for many years, in both theoretical and application aspects - the latter includes a variety of disciplines such as medicine, molecular biology, bioinformatics, genomics, proteomics, biometrics, security and manufacturing, just to mention a few. The essence of a PR system is to devise and evaluate an efficient classifier which is able to accurately classify unknown objects. Whether or not the classes are known, the recognition process is called supervised or unsupervised. In any case, devising a classifier requires the underlying objects be represented by means of features, typically, as multi-dimensional vectors, and the underlying classifiers are derived in terms of functions or algorithms on the corresponding multi-dimensional space. The main goals of this research proposal are to devise an efficient scheme for optimal clustering, which is capable of dealing with multi-dimensional data, and develop a robust linear dimensionality reduction scheme based on statistical properties of the classes, which is capable of achieving high accurate classification for a wide range of problems, while selecting the best subset of features. Applications of the newly proposed methods will be pursued on various problems in bioinformatics, proteomics, DNA microarrays and microbial biofilms. One of these is to efficiently cluster time-series DNA microarray data using unsupervised techniques. Another problem that will be tackled using the proposed clustering methods is DNA microarray image gridding and segmentation. Protein-protein interactions will be studied in terms of finding important residues and energetic factors in interfaces, and which are correlated to classification of protein complexes and amino acids.