Theam Foo Ng

Feature interaction in subspace clustering using the Choquet integral

Subspace clustering has recently emerged as a popular approach to removing irrelevant and redundant features during the clustering process. However, most subspace clustering methods do not consider the interaction between the features. This unawareness limits the analysis performance in many pattern recognition problems. In this paper, we propose a novel subspace clustering technique by introducing the feature interaction using the concepts of fuzzy measures and the Choquet integral. This new framework of subspace clustering can provide optimal subsets of interacted features chosen for each cluster, and hence can improve clustering-based pattern recognition tasks. Various experimental results illustrate the effective performance of the proposed method.

Impulse Noise Model and Its Variations

Impulse noise is one of the noise types that are normally corrupting digital images. In literature, impulse noises have been described in many ways. Therefore, the main aim of this manuscript is to survey a few impulse noise models that have been used by several researchers in digital image processing field. This will lead us to a better understanding on impulse noise models, and consequently will help the researchers to design more effective impulse noise reduction filters.

Automated Feature Weighting in Fuzzy Declustering-based Vector Quantization

Feature weighting plays an important role in improving the performance of clustering technique. We propose an automated feature weighting in fuzzy declustering-based vector quantization (FDVQ), namely AFDVQ algorithm, for enhancing effectiveness and efficiency in classification. The proposed AFDVQ imposes weights on the modified fuzzy c-means (FCM) so that it can automatically calculate feature weights based on their degrees of importance rather than treating them equally. Moreover, the extension of FDVQ and AFDVQ algorithms based on generalized improved fuzzy partitions (GIFP), known as GIFP-FDVQ and GIFP-AFDVQ respectively, are proposed. The experimental results on real data (original and noisy data) and modified data (biased and noisy-biased data) have demonstrated that the proposed algorithms outperformed standard algorithms in classifying clusters especially for biased data.

Awareness and Attitudes Towards Sustainable Development Amongst Higher Education Students in Penang, Malaysia

The role of youth in sustainable development decision-making and the implementation of sustainability programmes are critical elements to the long-term success of Agenda 21 and national sustainable agendas. Thus, advancing the role of youth and actively involving them in national sustainable agendas in the context of environmental protection and the promotion of economic and social development are crucially needed. However, there is inadequate information available about Malaysian youth’s awareness and attitudes with regard to this matter. The aim of this study is to determine the level of awareness and attitudes towards sustainable development amongst Malaysian youth. As an exploratory study, a survey was conducted in 2015 and 295 respondents from selected public and private higher education students in the state of Penang. This study has shown that the awareness of respondents about the concept and issues of sustainable development were well developed however, differed over semantics and what sustainable development encompasses. The survey also revealed that respondents were highly concerned about sustainability and were willing to practice more sustainable lifestyles. This study hopes to contribute as background information that will reflect on national sustainable development strategies.

Energy Consumption in Student Hostels of Universiti Sains Malaysia: Energy Audit and Energy Efficiency Awareness

Electricity is the main energy source consumed whereby our daily activities are predominantly and increasingly dependent on it. In the developing world where most countries steadily undergo rapid urbanization and population growth, energy consumption has immensely intensified over the last few decades. High energy demand is greatly propelled by consumption especially in residential, commercial, and university buildings. With the growing demands for electricity, increased costs of power and the desire to minimize dependence on energy generated by fossil fuels, initiative for efficient energy consumption particularly in building is given an increasing attention across the nation. Higher education is a growing sector with student numbers increasing every year. This means that the energy consumption of universities is also growing. This paper focuses on energy audit activities conducted in eight student hostel buildings of Universiti Sains Malaysia (USM) to assess the level of total energy consumption. Concurrently, the level of student engagement in sustainable energy consumption practices was evaluated. Furthermore, energy consumption for lighting in each hostel was determined through a survey. The results garnered from the audit were evaluated and feasible alternatives for potential energy saving and conservation measures were consequently identified and recommended to improve energy efficiency in the buildings.

Justification of Fuzzy Declustering Vector Quantization Modeling in Classification of Genotype‐Image Phenotypes

With the fast development of multi‐dimensional data compression and pattern classification techniques, vector quantization (VQ) has become a system that allows large reduction of data storage and computational effort. One of the most recent VQ techniques that handle the poor estimation of vector centroids due to biased data from undersampling is to use fuzzy declustering‐based vector quantization (FDVQ) technique. Therefore, in this paper, we are motivated to propose a justification of FDVQ based hidden Markov model (HMM) for investigating its effectiveness and efficiency in classification of genotype‐image phenotypes. The performance evaluation and comparison of the recognition accuracy between a proposed FDVQ based HMM (FDVQ‐HMM) and a well‐known LBG (Linde, Buzo, Gray) vector quantization based HMM (LBG‐HMM) will be carried out. The experimental results show that the performances of both FDVQ‐HMM and LBG‐HMM are almost similar. Finally, we have justified the competitiveness of FDVQ‐HMM in classification of cellular phenotype image database by using hypotheses t‐test. As a result, we have validated that the FDVQ algorithm is a robust and an efficient classification technique in the application of RNAi genome‐wide screening image data.

An Overview on the Application of Self-Adaptive Differential Evolution

Differential Evolution (DE) is possibly the most current powerful stochastic real-parameter optimization algorithm and has been used in multiple diverse area such as neural networks, logistics, scheduling, modelling and others. Its simplicity, ease of implementation and reliability had captures many practitioners and scientists in implementing the algorithm. As different problems require different parameter setting, the implementation of DE in tackling complex computational optimization problem is quite challenging. Nevertheless, success of the algorithm depends on the ability to choose the right parameter setting based on problems in hand. Thus, extra attention is needed in order to fine tune the perfect parameter for each problem. Self-adaptive Differential Evolution (SADE) algorithm had been introduced in order to simplify the search for the right parameter to be used in DE algorithm. With the introduction of SADE in optimization areas, where the choice of learning strategy and parameter setting do not require predefining, parameter tuning has become less confusing. This paper aims at providing an overview on significant application that have benefited from SADE implementation. SADE had been applied in numerous disciplines such as electromagnetics, power system, computer performance, fermentation, polyester process and more. SADE has also proven to achieve better performance compared to conventional DE algorithm. By collecting and analyzing related articles that have implemented SADE in solving problem, a significant trends on the application of SADE will be provided.

Learning lab on disaster risk management for sustainable development (DRM-SD): An evaluation

Purpose A series of “learning lab” projects on disaster risk management for sustainable development (DRM-SD) have been accomplished from 2014 to 2016 in Malaysia, Vietnam, Lao PDR and Cambodia by the Centre for Global Sustainability Studies. The project is designed for professionals from the disaster risk management field to encourage integration of sustainable development (SD) concerns into the larger planning framework for DRM. As a case study for capacity building (CB) evaluation, the central purpose of this study is to explore the approaches, feedbacks and implications of the DRM-SD CB project that have been developed and carried out. Design/methodology/approach Three methods have been used which are participation observations, surveys and document analysis. The results show that the project had successfully applied seven different tools to enhance analytical skills and professional knowledge of development practitioners in specific areas of DRM-SD. Findings Based on the survey, the project received positive response and valuable information from participants for future project development. Regarding the perspective of outcomes, the result indicates that south–south, ASEAN regional and triangular cooperation and role of higher education in DRM-SD are significant impacts from this project which can bring several benefits and should be promoted as an approach for the DRM-CB project as a whole. Originality/value It is hoped that this study will serve as a transfer learning initiative to provide approach guidelines and innovative mechanisms for DRM practitioners who will have the know-how and potential for leadership in DRM-SD.

Self-adapting approach in parameter tuning for differential evolution

Higher expectation has been requested from artificial intelligence (AI) owing to its success in various applications and domains. The use of AI is no longer limited to solve static optimization problems, but to perform well in dynamic optimization problems as well. The performance of AI in problem solving depends greatly on its own control parameters. The set of parameters which has been tuned to solve current optimization problem may not lead to the same performance if there is a shift or change in the optimization problem. To ensure its functionality in such condition, a machine learning needs to be able to self-determine its own control parameters. In short, a machine learning needs to be adaptive. Evolutionary algorithms (EAs) associated with adaptive ability turn out to be a potential solution under this condition. Therefore, our research focuses on the use of self-adaptive approach in parameter tuning in EAs, specifically in differential evolution (DE). Given that our proposed DE is no longer depending on a user to determine its control parameters, we are interested to know whether the self-adapting parameters will ensure good performance from DE or not. Two versions of DEs with the ability to self-adapt their parameters are developed. Most of DE related studies have suggested certain ranges of parameters to ensure appropriate operation of standard DE. In this research, we take the opportunity to confirm whether the ranges of self-adapting parameters fall within the suggested ranges or not. The experimental results have shown that both self-adapting DEs perform adequately well in 20 different benchmark problems without depending on user to determine the parameters explicitly. Besides that, it is interesting to find out the control parameters of the self-adapting DEs are not necessarily within the suggested ranges and they are still performing adequately well.

Self-adaptive differential evolution based on best and mean schemes

Differential evolution (DE), one of the evolutionary algorithms (EAs), is well-known for its quality solutions and speed convergence. Just like any EA, the execution of DE depends on the selection of its control parameters consisting of population size, crossover rate and scale factor. DE is operated by two different kinds of search mechanisms, i.e., exploration and exploitation. The selection of control parameters affects these search mechanisms, and thus, the performance of DE. Ranges on setting DEs control parameters are suggested in most studies but it still depends heavily on a users knowledge and experiences, as well as the types of problems. The common approach adopted by users is the trial-and-error method but this approach consumes time. Since adaptation has been responsible for the search optimal solutions in DE, the adaptability should be further utilized to determine DEs optimal control parameters. Therefore, we proposed a self-adaptive DE which is able to self-determine its control parameters based on an ensemble. An ensemble is operated based on two different parameter selection schemes, i.e., BEST and MEAN. The performances of DEs based on the selection schemes in 20 benchmarks problems are compared based on best-fitness, mean-fitness, crossover rate and scale factor. The experimental results showed that the proposed self-adaptive DEs are able to perform adequately well in the benchmark problems. Besides that, the results have shown an interesting pattern between crossover rate and scale factor when the DE is given freedom to determine its control parameters.