Maizura Mohd Sani

Support vector machine for classification of stress subjects using EEG signals

Stress is a mental condition that can effects the brain electrical activity to be different from the normal state. This brain cognitive change can be measured using EEG. The objective of this paper is to classify stress subjects based on EEG signal using SVM. The data which are used to represent stress subjects were taken from the residents of Pusat Darul Wardah; a shelter centre for troubled women. SVM is used to classify the EEG Alpha band data for Power Spectral Density and Energy Spectral Density. Using 5-fold cross validation, the classification rate are 83.33% for ESD data using RBF kernel function.

Prediction of time series based on load profile using JADE technique

This paper presents a method of mathematical technique using Joint Approximate Diagonalization of Eigen Matrices (JADE). The advantages of this method, it can predict the original analysis of source signal using limited information of the system as well as to minimize costs by reducing the number of equipment to estimate the type of load profile at certain locations. By using innovative methods based Power System Simulation (PSS™E) software, Phyton Programming, Matlab software and Microsoft Excel are implemented. The proposed technique is tested on 69 test bus system and several results such as simulation and error measures are also discussed in this paper.

A non linear face recognition system using Support Vector Machine

A face recognition system uses face to verify individuals using computing capability. However, its performances often degrade due to high dimensional data and large feature appearance of the face image. This paper present a face recognition system based on non linear feature extraction technique to reduce the dimensionality of the face image, called Locally Linear Embedding. This method considers the hidden layer of face manifold to be the input of a SVM multiclass classifier. The performance is evaluated using the ORL database and achieved better recognition rates than the Principal Component Analysis.