Title

Technical Papers Session I: Multi-dimension projection for non-linear data via Spearman correlation analysis (MD-SCA)

Abstract/Description

This paper introduces an algorithm of multidimensional informative projection or view of multiple variable and more than two random variables via Spearman correlation analysis (SCA). The proposed algorithm is an extension of Spearman correlation analysis to extract linear or nonlinear information of projections through pairwise correlation analysis. These multi-dimensional informative projections used as common patterns in pattern recognition application. The proposed algorithm extends SCA through linear algebraic solution for the optimization problem, the problem of dual representation of high multi-dimensional data, and structural dilemma issues along with deep learning model. Additionally, the proposed method decreases the quadratic algorithm complexity among linear and non-linear data through Spearman rank ability. The demonstration of proposed approached performs on two-bench mark data set: Face96 and Yale Face Database.

Location

Lecture Hall A (Aman Tower, 12th floor)

Session Theme

Technical Papers Session I - Data Science

Session Type

Parallel Technical Session

Session Chair

Dr Shahid Shaikh

Start Date

16-11-2019 3:10 PM

End Date

16-11-2019 3:30 PM

Share

COinS
 
Nov 16th, 3:10 PM Nov 16th, 3:30 PM

Technical Papers Session I: Multi-dimension projection for non-linear data via Spearman correlation analysis (MD-SCA)

Lecture Hall A (Aman Tower, 12th floor)

This paper introduces an algorithm of multidimensional informative projection or view of multiple variable and more than two random variables via Spearman correlation analysis (SCA). The proposed algorithm is an extension of Spearman correlation analysis to extract linear or nonlinear information of projections through pairwise correlation analysis. These multi-dimensional informative projections used as common patterns in pattern recognition application. The proposed algorithm extends SCA through linear algebraic solution for the optimization problem, the problem of dual representation of high multi-dimensional data, and structural dilemma issues along with deep learning model. Additionally, the proposed method decreases the quadratic algorithm complexity among linear and non-linear data through Spearman rank ability. The demonstration of proposed approached performs on two-bench mark data set: Face96 and Yale Face Database.