Research Review: A Novel Credit-based Harmony Search Algorithm
1030 until 1200
Viva Room 7th Floor
Mohd Khaled Yousef Shambour
Prof. Dr. Ahamad Tajudin Khader
Harmony search (HS) algorithm is a meta-heuristic algorithm which mimics the improvisation process of musical players when they try to reach a pleasant harmony. HS is used to tackle combinatorial problems efficiently which are hard to solve in terms of computational complexity. The random search and pitch adjustment operators in HS work on tuning the solution vector to avoid getting stuck in local minima. Selection mechanism is another factor, which helps in avoiding premature convergence caused by losing the diversity of population in the Harmony Memory (HM). On the other hand, the proper use of the solution vectors in HM (parents) when generating new harmony (child) affects the quality of the obtained solution. In this study, a new technique is proposed in which a different process of updating the HM is applied. A credit reward is added after each iteration to the parents of the child based on the contribution to the goodness of the child fitness. During updating the HM the parent with the worst accumulated credit is subjected to be replaced with the new generated harmony giving more chances for parents with better credit to survive. The proposed technique is applied to two different cases of optimization problems (i.e.unconstrained and constrained optimization problems). The experimental results show a considerable performance in favor of the proposed technique in terms of solution quality.
Research Review: Adapting and Enhancing Mussels Wandering Optimization Algorithm for Supervised-Training of Neural Networks
1500 until 1600
Meeting Room 7th Floor
Ahmed A. A. Abusnaina
Prof. Rosni Abdullah
Neural network (NN) uses mathematical models for information processing in order to accomplish a variety of tasks, which have been widely applied in engineering and sciences for diversity of applications. NNs are classified according to their computational units (neurons) as three generations: 1st:Networks based on McCulloch neuron, 2nd:Artificial Neural Networks (ANN) and 3rd:Spiking Neural Networks (SNN).
Networks of spiking neurons have been considered more biologically realistic and more powerful than their non-spiking predecessors as they can encode temporal information in their signals. Also, SNN differs in information processing inside its basic computational units “neurons” and the network structure by adding delay synapses between its connections.
The training process of NN deals with adjusting and altering the weights and/or structure of the network depending on a specific training algorithm. Training of NN fall into two main categories: traditional training algorithms and evolutionary-based training algorithms. Traditional training algorithms have several drawbacks such as local minima, training oscillation, and its slowness. Therefore, evolutionary-based training algorithms that depend on global optimization methods are utilized to train NN to overcome the drawbacks of traditional learning algorithms.
New evolutionary-based methods for training NN are proposed. These methods are based on adapting and enhancing the Mussels Wandering Optimization (MWO) algorithm for supervised-training of both ANN and SNN.
Research Review: Software Maintenance Expert based Decision support Model: A Knowledge Management Perspective
1530 until 1630
Meeting Room 7th Floor
Rahma A. Kamaludeen
Assoc. Prof. Dr. Cheah Yu-N.
Software maintenance is an ongoing process during the lifetime of an information system. Once it is on production the information system must evolve due to the operating environment changes, when new anomalies uncovered or when new user requirement arises. Prior to any software evolution, decision-making need to be undertaken by the organization’s IT decision makers whether to proceed with the evolution required, if yes on how to proceed with the required maintenance. However, the study on decision making in software maintenance is very limited, existing research focus mainly on project management. The uniqueness of this study relies on the ability of knowledge engineering method to capture the software maintenance knowledge of an information system that is inherent in the software expert’s mind, documentation and code. This uses the concept of expert system to record the knowledge of the information system in a knowledge base and inference logic to analyst the recorded knowledge. The analysis is carried out using inference logic pre-programmed into the expert-based Model. This is crucial especially to assist IT decision makers in determining whether it is possible to incorporate an additional requirement into a running system within an acceptable time-frame without the presence of a software expert in question. The Model is tested by proof of concept with a real life application of an organization together with case study evaluation conducted with IT decision makers of Malaysian industry.
Research Review: Profiling Malaysian ICT Academics on Their Choices of Commercialization Approches
1500 until 1630
Meeting Room 7th Floor
Prof. Dr. Rosni Abdullah
Dr. Nasriah Zakaria
The objective of this research is to profile the academics by exploring the micro level perspective which is the academics’ perceptions and experiences on their choices of commercialization approaches based on their personal characteristics. This is a grounded theory research project involves empirical setting of a multi-site case study carried out at five Malaysian research universities. The transcription and coding of interviews with the total of 42 academics led to the use of major and minor codes to develop a mid-range theory to encapsulate their choices of commercialization approaches into a coherent and abstract format.
The diversity personal characteristics along with the strength of supportive resources represent the intensity of the capability of an academic to act or react in terms of what research to be involved and what role to play in their academic careers. Results of this study suggest the profiling of the academics into five categories. First is the teaching academics who adopt single role and with very minimum research and no commercialization output. The second is pessimist-research academics who adopt dual role identity with fundamental research and no commercialization output. The third is the positivist-research academics who adopt applied research and choose to commercialize through external approaches. The fourth is the entrepreneurial academics who adopt applied research and choose to commercialize through internal approaches. The last one is the pure entrepreneurial academics who are persistent in effort to commercialization through quasi-internal approaches.
The findings are expected to shed new light on the way in which the academics choose their commercialization approaches. Therefore, it is expected to an insight view to the practitioners (industry, government and university’s management such as the Deans and TTO) where knowing specific characteristics of an academic would help them to promote such commercialization efforts.