Norazam Arbin

Comparative Analysis between K-Means and K-Medoids for Statistical Clustering

Clustering dynamic data is a challenge inidentifying and forming groups. This unsupervised learningusually leads to undirected knowledge discovery. The clusterdetection algorithm searches for clusters of data which aresimilar to one another by using similarity measures.Determining the suitable algorithm which can bring theoptimized groups cluster could be an issue. Depending on theparameters and attributes of the data, the results yielded fromusing both K-Means and K-Medoids could be varied. Thispaper presents a comparative analysis of both algorithms indifferent data clusters to lay out the strengths and weaknessesof both. Thorough studies were conducted in determining thecorrelation of the data with the algorithms to find therelationship among them.

Multi-objective Genetic Algorithm in Solving Conflicted Goals for Questions Generating Problem

Multi-objective genetic algorithm (MOGA) has been used for more than a decade to solve real-world optimization problems that have several, and often conflicting objectives. In this research, the conflicting objectives of achieving the maximum accuracy of the solution and at the same time minimizing the redundancy of the optimal solutions in retrieving the best set of exam questions for academicians for a particular subject are highlighted. Hence, the aim of this paper is to solve the multi-objective problem in a chromosome (solution) and also to maintain the fitness of the chromosome. The results of this research are measured based on the similarity achieved between the obtained and desired solutions. By using MOGA, a promising result is obtained with the maximum accuracy and simultaneously, minimizing the redundancy of the genes in a solution.

Marangoni convection in a micropolar fluid with feedback control

The effect of feedback control on the criterion for the onset of Marangoni convection in a horizontal micropolar fluid layer is studied theoretically. The upper free surface is assumed to be non-deformable and the lower boundary is taken to be rigid and isothermal with fixed temperature and span-vanishing boundaries. A linear stability analysis is used and the Galerkin method is employed to find the critical stability parameters numerically. It is found that the onset of instability can be delayed through the use of feedback control.

Optimizing Cluster of Questions by Using Dynamic Mutation in Genetic Algorithm

Clustering dynamic data is a challenge in identifying and forming groups. This unsupervised learning usually leads to indirect knowledge discovery. The cluster detection algorithm searches for clusters of data which are similar to one another by using similarity measures.Optimizing the clustered data with certain fixed values could be an issue. Depending on the parameters and attributes of the data, the results yielded probably either stuck in local optima or bias by attributes pattern. Performing Genetic Algorithm in the data cluster may increase the probability of the questions being clustered in the optimal group cluster. Dynamic Mutation in Genetic Algorithm used as repair mechanism to ensure the cluster is optimized enough and produce optimum indexed questions set.