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.
Keywords
Multi-Dimensional Projection, SqueezeNet, Spearman Correlation Analysis, Linear Algebraic Formulation
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
Recommended Citation
Khokhar, M. S., Cheng, K., Ayoub, M., Jamali, Z., & Eric, L. K. (2019). Technical Papers Session I: Multi-dimension projection for non-linear data via Spearman correlation analysis (MD-SCA). International Conference on Information and Communication Technologies. Retrieved from https://ir.iba.edu.pk/icict/2019/2019/8
COinS
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.