Zuriani Mustaffa

Enhanced artificial bee colony for training least squares support vector machines in commodity price forecasting

Abstract The importance of optimizing machine learning control parameters has motivated researchers to investigate for proficient optimization techniques. In this study, a Swarm Intelligence approach, namely artificial bee colony (ABC) is utilized to optimize parameters of least squares support vector machines. Considering critical issues such as enriching the searching strategy and preventing over fitting, two modifications to the original ABC are introduced. By using commodities prices time series as empirical data, the proposed technique is compared against two techniques, including Back Propagation Neural Network and by Genetic Algorithm. Empirical results show the capability of the proposed technique in producing higher prediction accuracy for the prices of interested time series data.

Dengue Outbreak Prediction: A Least Squares Support Vector Machines Approach

Dengue fever (DF) and the potentially fatal dengue hemorrhagic fever (DHF) continue to be a crucial public health concern in Malaysia. This paper proposes a prediction model that incorporates Least Squares Support Vector Machines (LS-SVM) in predicting future dengue outbreak. Data sets used in the undertaken study includes data on dengue cases and rainfall level collected in five districts in Selangor. Data were preprocessed using the Decimal Point Normalization before being fed into the training model. Prediction results of unseen data show that the LS-SVM prediction model outperformed the Neural Network model in terms of prediction accuracy and computational time.

Training LSSVM with GWO for price forecasting

This paper presents a hybrid forecasting model namely Grey Wolf Optimizer-Least Squares Support Vector Machines (GWO-LSSVM). In this study, a great deal of attention was paid in determining LSSVMs hyper parameters. For that matter, the GWO is utilized an optimization tool for optimizing the said hyper parameters. Realized in gold price forecasting, the feasibility of GWO-LSSVM is measured based on Mean Absolute Percentage Error (MAPE) and Root Mean Square Percentage Error (RMSPE). Upon completing the simulation tasks, the comparison against two hybrid methods suggested that the GWO-LSSVM capable to produce lower forecasting error as compared to the identified forecasting techniques.

Optimal Reactive Power Dispatch Solution by Loss Minimization Using Moth-Flame Optimization Technique

Abstract In this paper, a newly surfaced nature-inspired optimization technique called moth-flame optimization (MFO) algorithm is utilized to address the optimal reactive power dispatch (ORPD) problem. MFO algorithm is inspired by the natural navigation technique of moths when they travel at night, where they use visible light sources as guidance. In this paper, MFO is realized in ORPD problem to investigate the best combination of control variables including generators voltage, transformers tap setting as well as reactive compensators sizing to achieve minimum total power loss and minimum voltage deviation. Furthermore, the effectiveness of MFO algorithm is compared with other identified optimization techniques on three case studies, namely IEEE 30-bus system, IEEE 57-bus system and IEEE 118-bus system. The statistical analysis of this research illustrated that MFO is able to produce competitive results by yielding lower power loss and lower voltage deviation than the selected techniques from literature.

Forecasting Model Based on LSSVM and ABC for Natural Resource Commodity

 Abstract—Reliable forecasts of the price of natural resource commodity is of interest for a wide range of applications. This includes generating macroeconomic projections and in assessing macroeconomic risks. Various approaches have been introduced in developing the required forecasting models. In this paper, a forecasting model based on an optimized Least Squares Support Vector Machine is proposed. The determination of hyper-parameters is performed using a nature inspired algorithm, Artificial Bee Colony. The proposed forecasting model is realized in gold price forecasting. The undertaken experiments showed that the prediction accuracy and Mean Absolute Percentage Error produced by the proposed model is better compared to the one produced using Least Squares Support Vector Machine as an individual.

LS-SVM hyper-parameters optimization based on GWO algorithm for time series forecasting

The importance of optimizing Least Squares Support Vector Machines (LSSVM) embedded control parameters has motivated researchers to search for proficient optimization techniques. In this study, a new metaheuristic algorithm, viz., Grey Wolf Optimizer (GWO), is employed to optimize the parameters of interest. Realized in commodity time series data, the proposed technique is compared against two comparable techniques, including single GWO and LSSVM optimized by Artificial Bee Colony (ABC) algorithm (ABC-LSSVM). Empirical results suggested that the GWO-LSSVM is capable to produce lower error rates as compared to the identified algorithms for the price of interested time series data.

Application of LSSVM by ABC in energy commodity price forecasting

The importance of the hyper parameters selection for a kernel-based algorithm, viz. Least Squares Support Vector Machines (LSSVM) has been a critical concern in literature. In order to meet the requirement, this work utilizes a variant of Artificial Bee Colony (known as mABC) for hyper parameters selection of LSSVM. The mABC contributes in the exploitation process of the artificial bees and is based on Levy mutation. Realized in crude oil price forecasting, the performance of mABC-LSSVM is guided based on Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSPE) and compared against the standard ABC-LSSVM and LSSVM optimized by Genetic Algorithm. Empirical results suggested that the mABC-LSSVM is superior than the chosen benchmark algorithms.

Developing a gold price predictive analysis using Grey Wolf Optimizer

As the value of gold cannot be blindly rejected, forecasting the future prices of gold has long been an intriguing topic and is extensively studied by researchers from different fields including economics, statistics, and computer science. The motivation for these studies is naturally to predict the future prices so that gold can be bought and sold at profitable positions and reduce the risk of investment. However, there are still a lot of untackled questions and room for improvements in these forecasting techniques. This is because there are no optimal models for all forecasting problems. Different question needs a different answer; therefore, more experiments and modeling need to be done in order for researcher to enhance their findings. The target of this paper is to present a gold forecasting techniques using one of the optimization algorithm called Grey Wolf Optimizer (GWO).

Block-based Tchebichef image watermarking scheme using psychovisual threshold

Digital multimedia has drastically increased the production and distribution of digital data in the recent years. Unauthorized manipulation and ownership of digital image have become a serious issue. In this paper, we propose a watermarking scheme which uses block-based Tchebichef moments considering psychovisual threshold. The psychovisual threshold is used to prescribe the potential location of embedded watermark. The proposed watermarking scheme considers minimum modified entropy values to determine the embedded blocks. The lowest psychovisual error threshold on each selected block are chosen as the best location to insert the watermark image. Experimental results demonstrate that the embedding watermark into the lowest Tchebichef psychovisual threshold can produce a good level of imperceptibility. The watermark recovery is strongly robust against JPEG compression.

Cuckoo Search Algorithm as an optimizer for Optimal Reactive Power Dispatch problems

This paper presents the application of Cuckoo Search Algorithm (CSA) in optimizing the control variables of power system operation in solving the optimal reactive power dispatch (ORPD) problem. CSA is inspired by the parasitic behavior of Cuckoo birds in reproduction process based on the probability for a host bird in discovering an alien egg in its nest. The implementation of CSA in determining the optimal value of control variables such as generator bus voltages, transformer tap setting and shunt reactive elements in order to obtain the minimize loss in the system. In this paper, IEEE-30 bus system is utilized to show the effectiveness of CSA and then the comparison with other nature inspired algorithms will be presented.