Ruhaidah Samsudin

A hybrid model of self-organizing maps (SOM) and least square support vector machine (LSSVM) for time-series forecasting

Support vector machine is a new tool from Artificial Intelligence (AI) field has been successfully applied for a wide variety of problem especially in time-series forecasting. In this paper, least square support vector machine (LSSVM) is an improved algorithm based on SVM, with the combination of self-organizing maps(SOM) also known as SOM-LSSVM is proposed for time-series forecasting. The objective of this paper is to examine the flexibility of SOM-LSSVM by comparing it with a single LSSVM model. To assess the effectiveness of SOM-LSSVM model, two well-known datasets known as the Wolf yearly sunspot data and the Monthly unemployed young women data are used in this study. The experiment shows SOM-LSSVM outperforms the single LSSVM model based on the criteria of mean absolute error (MAE) and root mean square error (RMSE). It also indicates that SOM-LSSVM provides a promising alternative technique in time-series forecasting.

Crude Oil Price Forecasting Based on Hybridizing Wavelet Multiple Linear Regression Model, Particle Swarm Optimization Techniques, and Principal Component Analysis

Crude oil prices do play significant role in the global economy and are a key input into option pricing formulas, portfolio allocation, and risk measurement. In this paper, a hybrid model integrating wavelet and multiple linear regressions (MLR) is proposed for crude oil price forecasting. In this model, Mallat wavelet transform is first selected to decompose an original time series into several subseries with different scale. Then, the principal component analysis (PCA) is used in processing subseries data in MLR for crude oil price forecasting. The particle swarm optimization (PSO) is used to adopt the optimal parameters of the MLR model. To assess the effectiveness of this model, daily crude oil market, West Texas Intermediate (WTI), has been used as the case study. Time series prediction capability performance of the WMLR model is compared with the MLR, ARIMA, and GARCH models using various statistics measures. The experimental results show that the proposed model outperforms the individual models in forecasting of the crude oil prices series.

A hybrid GMDH and least squares support vector machines in time series forecasting

Time series consists of complex nonlinear and chaotic patterns that are difficult to forecast. This paper proposes a novel hybrid forecasting model which combines the group method of data handling (GMDH) and the least squares support vector machine (LSSVM), known as GLSSVM. The GMDH is used to determine the useful input variables for the LSSVM model and the LSSVM model that works as time series forecasting. Three well-known time series data sets are used in this study to demonstrate the effectiveness of the forecasting model. These data are utilized to forecast through an application aimed to handle real life time series. The results found by the proposed model were compared with the results of the GMDH and LSSVM models. Experiment result indicates that the hybrid model was a powerful tool to model time series data and provides a promising technique in time series forecasting methods.

Daily crude oil price forecasting using hybridizing wavelet and artificial neural network model

A new method based on integrating discrete wavelet transform and artificial neural networks (WANN) model for daily crude oil price forecasting is proposed. The discrete Mallat wavelet transform is used to decompose the crude price series into one approximation series and some details series (DS). The new series obtained by adding the effective one approximation series and DS component is then used as input into the ANN model to forecast crude oil price. The relative performance of WANN model was compared to regular ANN model for crude oil forecasting at lead times of 1 day for two main crude oil price series, West Texas Intermediate (WTI) and Brent crude oil spot prices. In both cases, WANN model was found to provide more accurate crude oil prices forecasts than individual ANN model.

Fishery Landing Forecasting Using Wavelet-Based Autoregressive Integrated Moving Average Models

The accuracy of the wavelet-ARIMA (WA) model in monthly fishery landing forecasting is investigated in the study. In the first part of the study, the discrete wallet transform (DWT) is used to decompose fishery landing time series data. Then ARIMA, as a powerful forecasting tool, is implemented to predict each wavelet transform subseries components independently. Finally, the prediction results of the modeled subseries components are summed to formulate an ensemble forecast for the original fishery landing series. To assess the effectiveness of this model, monthly fishery landing recorded data from East Johor and Pahang states of Peninsular Malaysia have been used as a case study. The result of the study shows that the proposed model was found to provide more accurate fishery landing series forecasts than the individual ARIMA model.

Hybridizing Wavelet and Multiple Linear Regression Model for Crude Oil Price Forecasting

Crude oil prices play a significant role in the global economy and contribute an important factor affecting government’s plans and commercial sectors. In this paper, the accuracy of the simple wavelet multiple linear regression (WMLR) model in crude oil prices forecasting was investigated. The WMLR model was improved by combining two methods: discrete wavelet transform (DWT) and a multiple linear regression (MLR) model. To assess the effectiveness of this model, daily crude oil market-West Texas Intermediate (WTI) was used as the case study. Time series prediction capability performance of the WMLR model is compared with the Artificial neural network (ANN), autocorrelation integrated moving average (ARIMA), MLR and Generalized Autoregressive Conditional Heteroscedasticity (GARCH) models using various statistics measures. The results show that the hybrid WMLR is more accurate and perform better than of any individual model in the prediction of crude oil prices series.

Comparison of Drought Forecasting Using ARIMA and Empirical Wavelet Transform-ARIMA

Drought forecasting is important in preparing for drought and its mitigation plan. This paper focuses on the investigation of ARIMA and Empirical Wavelet Transform (EWT)-ARIMA in forecasting drought using Standard Precipitation Index (SPI). EWT is employed to decompose the time series into 4 modes. SPI of 3, 6, 9, 12 and 24 months were used. The objective of this study is to compare the effectiveness of the methods in accurately forecasting drought in Arau, Malaysia. It was found that EWT-ARIMA perform better when SPI is 3, 6 or 24. EWT-ARIMA performs comparably to ARIMA when the SPI is 9 or 12.

A Review of Crude Oil Prices Forecasting using Hybrid Method

Received May 20, 2018 Revised Jun 21, 2018 Accepted Jun 25, 2018 RFID technology is a Radio frequency identification system that provides a reader reading the data item from its tag. Nowadays, RFID system has rapidly become more common in our life because of its autonomous advantages compared to the traditional barcode. It can detect hundreds of tagged items automatically at a time. However, in RFID, missing tag detection can occur due to signal collisions and interferences. It will cause the system to report incorrect tag’s count due to an incorrect number of tags being detected. The consequences of this problem can be enormous to business, as it will cause incorrect business decisions to be made. Thus, a Missing Tag Detection Algorithm (MTDA) is proposed to solve the missing tag detection problem. There are many other existing approaches has been proposed including Window Sub-range Transition Detection (WSTD), Efficient Missing-Tag Detection Protocol (EMD) and Multi-hashing based Missing Tag Identification (MMTI) protocol. The result from experiments shows that our proposed approach performs better than the other in terms of execution time and reliability.Received May 1, 2018 Revised Jun 21, 2018 Accepted Jun 28, 2018 The diagnosing features for Diabetic Retinopathy (DR) comprises of features occurring in and around the regions of blood vessel zone which will result into exudes, hemorrhages, microaneurysms and generation of textures on the albumen region of eye balls. In this study we presenta probabilistic convolution neural network based algorithms, utilized for the extraction of such features from the retinal images of patient’s eyeballs. The classificat ions proficiency of various DR systems is tabulated and examined. The majority of the reported systems are profoundly advanced regarding the analyzed fundus images is catching up to the human ophthalmologist’s characterization capacities.Received Nov 25, 2017 Revised Jan 9, 2018 Accepted May 27, 2018 The usage of multilevel inverter has increased in a drastic manner for the past years. These novel inverters are useful in various mega power applications. As they are having the ability to change the output waveforms, they are having good harmonic distortions and better output results. This work proposes a novel five level asymmetrical inverter which is incorporated with the zeta converter. Comparison is made with the existing multilevel inverter with the proposed system. The simulation results give the proposed system has less THD [1] when compared to the existing multilevel inverters. The main objective is that the number of switches and capacitors are reduced which in turn reduces the loss and the cost. From the output results is has been proved that the proposed topology gives reduced loss and high quality output when compared with the conventional methods.Received Dec 26, 2017 Revised Jan 09, 2018 Accepted May 26, 2018 Evolutionary Algorithms (EAs) are the potential tools for solving optimization problems. The EAs are the population based algorithms and they search for the optimal solution(s) from a initial set of candidates solutions known as population. This population is to be initialized at first before the evolution of the algorithm starts. There exists different ways to initialize this population. Understanding and choosing the right population initialization technique for the given problem is a difficult task for the researchers and problem solvers. To alleviate this issue, this paper is framed with two objectives. The first objective is to present the details of various Population Initialization (PI) techniques of EAs, for the readers to give brief description of all the PI techniques. The second objective is to present the steps and empirical comparison of the results of two different PI techniques implemented for Differential Evolution (DE) algorithm. Theoretical insights and empirical results of the PI techniques are presented in this paper.Received May 23, 2018 Revised Jun 21, 2018 Accepted Jul 2, 2018 Smoothing filters are essential for noise removal and image restoration. Gaussian filters are used in many digital image and video processing systems. Hence the hardware implementation of the Gaussian filter becomes a reliable solution for real time image processing applications. This paper discusses the implementation of a novel Gaussian smoothing filter with low power approximate adders in Field Programmable Gate Array (FPGA). The proposed Gaussian filter is applied to restore the noisy images in the proposed system. Original test images with 512x512 pixels were taken and divided in to 4x4 blocks with 256x256 pixels. The proposed technique has been applied and the performance metrics were measured for various simulation criteria. The proposed algorithm is also implemented using approximate adders, since approximate adders had been recognized as a reliable alternate for error tolerant applications in circuit based metrics such as power, area and delay where the accuracy may be considered for trade off.Muhammad Farrel Pramono 1 , Kevin Renalda 2 , Harco Leslie Hendric Spits Warnars 3 , Dedy Prasetya Kristiadi 4 , Worapan kusakunniran 1,2 Information Systems Department, School of Information Systems, Bina Nusantara University, Jakarta, Indonesia 11480 3 Computer Science Department, BINUS Graduate Program-Doctor of Computer Science, Bina Nusantara University , Jakarta, Indonesia 11480 4 Computer Systems, STMIK Raharja, Tangerang Banten, Indonesia 15119 Faculty of Information and Communication Technology, Mahidol University, Nakhon Pathom, ThailandSoft Computing and Data Mining Centre, Faculty of Computer Sciences and Information Technology, Universiti Tun Hussein Onn Malaysia Faculty of Creative Technology and Heritage, Universiti Malaysia Kelantan, Karung Berkunci 01, 16300, Bachok, Kelantan, Malaysia School of Industrial Engineering, Telkom University, 40257 Bandung, West Java, Indonesia Laboratory of Biodiversity and Bioinformatics, Universiti Teknologi Malaysia, 81300 Skudai, Johor, Malaysia Faculty of Computer Systems and Software Engineering, Universiti Malaysia Pahang, Pahang, Malaysia Department of Software Engineering & Information System, Faculty of Computer Science and Information Technology, University Putra Malaysia (UPM), 43400 Selangor, Serdang, MalaysiaReceived Mar 19, 2018 Revised May 20, 2018 Accepted Jun 3, 2018 Kinect-based physical rehabilitation grows significantly as a mechanism for clinical assessment and rehabilitation due to its flexibility, low-cost and markerless system for human action capture. It is also an approach to provide convenience for for patients’ exercises continuation at home. In this paper, we discuss a review of the present Kinect-based physiotherapy and assessment for rehabilitation patients to provide an outline of the state of art, limitation and issues of concern as well as suggestion for future work in this approach. The paper is constructed into three main parts. The introduction was discussed on physiotherapy exercises and the limitation of current Kinect-based applications. Next, we also discuss on Kinect Skeleton Joint and Kinect Depth Map features that being used widely nowadays. A concise summary with significant findings of each paper had been tabulate for each feature; Skeleton Joints and Depth Map. Afterwards, we assemble a quite number of classification method that being implemented for activity recognition in past few years.Received May 9, 2018 Revised Jun 2, 2018 Accepted Jun 21, 2018 As the cloud computing is gaining more user base the problem of simultaneously catering computational resources to multitude of users or their application is on rise. It remains a critical problem and pose hindrance in scalability of cloud computing. Thus, in order to layout the proper solution for the mentioned problem; it is necessary to sum up a proper knowledge based of the existing solution, there drawbacks and a detail analysis of its performances. In this study we present a review of multi-tenant frameworks and approaches used in the industry which reaps advantages to facilitate multi-tenancy.

Forecasting Drought Using Modified Empirical Wavelet Transform-ARIMA with Fuzzy C-Means Clustering

Received May 20, 2018 Revised Jun 21, 2018 Accepted Jun 25, 2018 RFID technology is a Radio frequency identification system that provides a reader reading the data item from its tag. Nowadays, RFID system has rapidly become more common in our life because of its autonomous advantages compared to the traditional barcode. It can detect hundreds of tagged items automatically at a time. However, in RFID, missing tag detection can occur due to signal collisions and interferences. It will cause the system to report incorrect tag’s count due to an incorrect number of tags being detected. The consequences of this problem can be enormous to business, as it will cause incorrect business decisions to be made. Thus, a Missing Tag Detection Algorithm (MTDA) is proposed to solve the missing tag detection problem. There are many other existing approaches has been proposed including Window Sub-range Transition Detection (WSTD), Efficient Missing-Tag Detection Protocol (EMD) and Multi-hashing based Missing Tag Identification (MMTI) protocol. The result from experiments shows that our proposed approach performs better than the other in terms of execution time and reliability.Received May 1, 2018 Revised Jun 21, 2018 Accepted Jun 28, 2018 The diagnosing features for Diabetic Retinopathy (DR) comprises of features occurring in and around the regions of blood vessel zone which will result into exudes, hemorrhages, microaneurysms and generation of textures on the albumen region of eye balls. In this study we presenta probabilistic convolution neural network based algorithms, utilized for the extraction of such features from the retinal images of patient’s eyeballs. The classificat ions proficiency of various DR systems is tabulated and examined. The majority of the reported systems are profoundly advanced regarding the analyzed fundus images is catching up to the human ophthalmologist’s characterization capacities.Received Nov 25, 2017 Revised Jan 9, 2018 Accepted May 27, 2018 The usage of multilevel inverter has increased in a drastic manner for the past years. These novel inverters are useful in various mega power applications. As they are having the ability to change the output waveforms, they are having good harmonic distortions and better output results. This work proposes a novel five level asymmetrical inverter which is incorporated with the zeta converter. Comparison is made with the existing multilevel inverter with the proposed system. The simulation results give the proposed system has less THD [1] when compared to the existing multilevel inverters. The main objective is that the number of switches and capacitors are reduced which in turn reduces the loss and the cost. From the output results is has been proved that the proposed topology gives reduced loss and high quality output when compared with the conventional methods.Received Dec 26, 2017 Revised Jan 09, 2018 Accepted May 26, 2018 Evolutionary Algorithms (EAs) are the potential tools for solving optimization problems. The EAs are the population based algorithms and they search for the optimal solution(s) from a initial set of candidates solutions known as population. This population is to be initialized at first before the evolution of the algorithm starts. There exists different ways to initialize this population. Understanding and choosing the right population initialization technique for the given problem is a difficult task for the researchers and problem solvers. To alleviate this issue, this paper is framed with two objectives. The first objective is to present the details of various Population Initialization (PI) techniques of EAs, for the readers to give brief description of all the PI techniques. The second objective is to present the steps and empirical comparison of the results of two different PI techniques implemented for Differential Evolution (DE) algorithm. Theoretical insights and empirical results of the PI techniques are presented in this paper.Received May 23, 2018 Revised Jun 21, 2018 Accepted Jul 2, 2018 Smoothing filters are essential for noise removal and image restoration. Gaussian filters are used in many digital image and video processing systems. Hence the hardware implementation of the Gaussian filter becomes a reliable solution for real time image processing applications. This paper discusses the implementation of a novel Gaussian smoothing filter with low power approximate adders in Field Programmable Gate Array (FPGA). The proposed Gaussian filter is applied to restore the noisy images in the proposed system. Original test images with 512x512 pixels were taken and divided in to 4x4 blocks with 256x256 pixels. The proposed technique has been applied and the performance metrics were measured for various simulation criteria. The proposed algorithm is also implemented using approximate adders, since approximate adders had been recognized as a reliable alternate for error tolerant applications in circuit based metrics such as power, area and delay where the accuracy may be considered for trade off.Muhammad Farrel Pramono 1 , Kevin Renalda 2 , Harco Leslie Hendric Spits Warnars 3 , Dedy Prasetya Kristiadi 4 , Worapan kusakunniran 1,2 Information Systems Department, School of Information Systems, Bina Nusantara University, Jakarta, Indonesia 11480 3 Computer Science Department, BINUS Graduate Program-Doctor of Computer Science, Bina Nusantara University , Jakarta, Indonesia 11480 4 Computer Systems, STMIK Raharja, Tangerang Banten, Indonesia 15119 Faculty of Information and Communication Technology, Mahidol University, Nakhon Pathom, ThailandSoft Computing and Data Mining Centre, Faculty of Computer Sciences and Information Technology, Universiti Tun Hussein Onn Malaysia Faculty of Creative Technology and Heritage, Universiti Malaysia Kelantan, Karung Berkunci 01, 16300, Bachok, Kelantan, Malaysia School of Industrial Engineering, Telkom University, 40257 Bandung, West Java, Indonesia Laboratory of Biodiversity and Bioinformatics, Universiti Teknologi Malaysia, 81300 Skudai, Johor, Malaysia Faculty of Computer Systems and Software Engineering, Universiti Malaysia Pahang, Pahang, Malaysia Department of Software Engineering & Information System, Faculty of Computer Science and Information Technology, University Putra Malaysia (UPM), 43400 Selangor, Serdang, MalaysiaReceived Mar 19, 2018 Revised May 20, 2018 Accepted Jun 3, 2018 Kinect-based physical rehabilitation grows significantly as a mechanism for clinical assessment and rehabilitation due to its flexibility, low-cost and markerless system for human action capture. It is also an approach to provide convenience for for patients’ exercises continuation at home. In this paper, we discuss a review of the present Kinect-based physiotherapy and assessment for rehabilitation patients to provide an outline of the state of art, limitation and issues of concern as well as suggestion for future work in this approach. The paper is constructed into three main parts. The introduction was discussed on physiotherapy exercises and the limitation of current Kinect-based applications. Next, we also discuss on Kinect Skeleton Joint and Kinect Depth Map features that being used widely nowadays. A concise summary with significant findings of each paper had been tabulate for each feature; Skeleton Joints and Depth Map. Afterwards, we assemble a quite number of classification method that being implemented for activity recognition in past few years.Received May 9, 2018 Revised Jun 2, 2018 Accepted Jun 21, 2018 As the cloud computing is gaining more user base the problem of simultaneously catering computational resources to multitude of users or their application is on rise. It remains a critical problem and pose hindrance in scalability of cloud computing. Thus, in order to layout the proper solution for the mentioned problem; it is necessary to sum up a proper knowledge based of the existing solution, there drawbacks and a detail analysis of its performances. In this study we present a review of multi-tenant frameworks and approaches used in the industry which reaps advantages to facilitate multi-tenancy.

Tourism forecasting using modified empirical mode decomposition and group method of data handling

In this study, a hybrid model using modified Empirical Mode Decomposition (EMD) and Group Method of Data Handling (GMDH) model is proposed for tourism forecasting. This approach reconstructs intrinsic mode functions (IMFs) produced by EMD using trial and error method. The new component and the remaining IMFs is then predicted respectively using GMDH model. Finally, the forecasted results for each component are aggregated to construct an ensemble forecast. The data used in this experiment are monthly time series data of tourist arrivals from China, Thailand and India to Malaysia from year 2000 to 2016. The performance of the model is evaluated using Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) where conventional GMDH model and EMD-GMDH model are used as benchmark models. Empirical results proved that the proposed model performed better forecasts than the benchmarked models.