Nor Azura Md. Ghani
Universiti Teknologi MARA

Published : 8 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 8 Documents
Search

Performance analysis of supervised learning models for product title classification Norsyela Muhammad Noor Mathivanan; Nor Azura Md. Ghani; Roziah Mohd Janor
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 8, No 3: September 2019
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (447.973 KB) | DOI: 10.11591/ijai.v8.i3.pp228-236

Abstract

Online business development through e-commerce platforms is a phenomenon which change the world of promoting and selling products in this 21st century. Product title classification is an important task in assisting retailers and sellers to list a product in a suitable category. Product title classification is a part of text classification problem but the properties of product title are different from general document. This study aims to evaluate the performance of five different supervised learning models on data sets consist of e-commerce product titles with a very short description and they are incomplete sentences. The supervised learning models involve in the study are Naïve Bayes, K-Nearest Neighbor (KNN), Decision Tree, Support Vector Machine (SVM) and Random Forest. The results show KNN model is the best model with the highest accuracy and fastest computation time to classify the data used in the study. Hence, KNN model is a good approach in classifying e-commerce products.
Exploration on digital marketing as business strategy model among Malaysian entrepreneurs via neurocomputing Hazrita Ab Rahim; Shafaf Ibrahim; Saadi Bin Ahmad Kamaruddin; Nor Azura Md. Ghani; Ismail Musirin
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 9, No 1: March 2020
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (337.058 KB) | DOI: 10.11591/ijai.v9.i1.pp18-24

Abstract

Artificial Intelligence is great when it comes to routine activities and vast amounts of data are analyzed. This can be done more quickly and efficiently than men. In the world of digital marketing, Artificial Intelligence is quickly coming into play. With Artificial Intelligence joining the digital marketing environment, predicting user behavior, search cycles, and much more will be easier. This can support websites that are highly user-friendly for organizations. Moreover, with the aid of Artificial Intelligence, content creation has become a faster and easier task for brands. Practically, a company's degree of enterprise marketing can have an effect on its overall business efficiency. Entrepreneurial marketing is driven by entrepreneurial opportunities which involves the proactive identification and exploitation of opportunities for acquiring and retaining profitable customers through Digital approaches to risk management, resource leveraging and value creation. This research was done by collecting data using semi structure questionnaire distributed to 169 start up owners in Klang Valley area. Using two-layer 6-3-1 with hyperbolic tangent-purelin configurations neural network model, it was found that proactiveness, risk taking, resource leveraging, opportunity focus, intensity and value add are the significant factors towards digital marketing respectively. It is expected that the findings would give some inputs to the Malaysian entrepreneurs on innovative digital marketing in their businesses, regardless the sizes.
The Application of Modified Least Trimmed Squares with Genetic Algorithms Method in Face Recognition Nur Azimah Abdul Rahim; Nor Azura Md. Ghani; Norazan Mohamed; Hishamuddin Hashim; Ismail Musirin
Indonesian Journal of Electrical Engineering and Computer Science Vol 8, No 1: October 2017
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v8.i1.pp154-158

Abstract

Severely occluded face images are the main problem in low performance of face recognition algorithms. In this paper, we apply a new algorithm, a modified version of the least trimmed squares (LTS) with a genetic algorithms introduce by [1]. We focused on the application of modified LTS with genetic algorithm method for face image recognition. This algorithm uses genetic algorithms to construct a basic subset rather than selecting the basic subset randomly. The modification in this method lessens the number of trials to obtain the minimum of the LTS objective function. This method was then applied to two benchmark datasets with clean and occluded query images. The performance of this method was measured by recognition rates. The AT&T dataset and Yale Dataset with different image pixel sizes were used to assess the method in performing face recognition. The query images were contaminated with salt and pepper noise. The modified LTS with GAs method is applied in face recognition framework by using the contaminated images as query image in the context of linear regression. By the end of this study, we can determine this either this method can perform well in dealing with occluded images or vice versa.
Active and Reactive Power Scheduling Optimization using Firefly Algorithm to Improve Voltage Stability under Load Demand Variation Mohamad Khairuzzaman Mohamad Zamani; Ismail Musirin; Halim Hassan; Sharifah Azwa Shaaya; Shahril Irwan Sulaiman; Nor Azura Md. Ghani; Saiful Izwan Suliman
Indonesian Journal of Electrical Engineering and Computer Science Vol 9, No 2: February 2018
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v9.i2.pp365-372

Abstract

This paper presents active and reactive power scheduling using firefly algorithm (FA) to improve voltage stability under load demand variation. The study involves the development of firefly optimization engine for power scheduling process involving the active and reactive power for wind generator. The scheduling optimization of wind generator is tested by using IEEE 30-Bus Reliability Test System (RTS). Voltage stability of the system is assessed based in a pre-developed voltage stability indicator termed as fast voltage stability index (FVSI). This study also considers the effects on the loss and voltage profile of the system resulted from the optimization, where the FVSI value at the observed line, minimum voltage of the system and loss were monitored during the load increment. Results obtained from the study are convincing in addressing the scheduling of power in wind generator. Implementation of FA approach to solve power scheduling revealed its flexibility and feasible for solving larger system within different objective functions.This paper presents active and reactive power scheduling using firefly algorithm (FA) to improve voltage stability under load demand variation. The study involves the development of firefly optimization engine for power scheduling process involving the active and reactive power for wind generator. The scheduling optimization of wind generator is tested by using IEEE 30-Bus Reliability Test System (RTS). Voltage stability of the system is assessed based in a pre-developed voltage stability indicator termed as fast voltage stability index (FVSI). This study also considers the effects on the loss and voltage profile of the system resulted from the optimization, where the FVSI value at the observed line, minimum voltage of the system and loss were monitored during the load increment. Results obtained from the study are convincing in addressing the scheduling of power in wind generator. Implementation of FA approach to solve power scheduling revealed its flexibility and feasible for solving larger system within different objective functions.
Enhanced BFGS Quasi-Newton Backpropagation Models on MCCI Data Nor Azura Md. Ghani; Saadi Ahmad Kamaruddin; Norazan Mohammed Ramli; Ismail Musirin; Hishamuddin Hashim
Indonesian Journal of Electrical Engineering and Computer Science Vol 8, No 1: October 2017
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v8.i1.pp101-106

Abstract

Neurocomputing is widely implemented in time series area, however the nearness of exceptions that for the most part happen in information time arrangement might be hurtful to the information organize preparing. This is on the grounds that the capacity to consequently discover any examples without earlier suppositions and loss of all-inclusive statement. In principle, the most well-known preparing calculation for Backpropagation calculations inclines toward lessening ordinary least squares estimator (OLS) or all the more particularly, the mean squared error (MSE). In any case, this calculation is not completely hearty when exceptions exist in preparing information, and it will prompt false estimate future esteem. Along these lines, in this paper, we show another calculation that control calculations firefly on slightest middle squares estimator (FFA-LMedS) for BFGS quasi-newton backpropagation neural network nonlinear autoregressive moving (BPNN-NARMA) model to lessen the effect of exceptions in time arrangement information. In the in the mean time, the monthly data of Malaysian Roof Materials cost index from January 1980 to December 2012 (base year 1980=100) with various level of exceptions issue is adjusted in this examination. Toward the finish of this paper, it was found that the upgraded BPNN-NARMA models utilizing FFA-LMedS performed extremely well with RMSE values just about zero errors. It is expected that the finding would help the specialists in Malaysian development activities to handle cost indices data accordingly.
Modified BPNN via Iterated Least Median Squares, Particle Swarm Optimization and Firefly Algorithm Nor Azura Md. Ghani; Saadi bin Ahmad Kamaruddin; Norazan Mohamed Ramli; Ismail Musirin; Hishamuddin Hashim
Indonesian Journal of Electrical Engineering and Computer Science Vol 8, No 3: December 2017
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v8.i3.pp779-786

Abstract

There is doubtlessly manufactured artificial neural system (ANN) is a standout amongst the most acclaimed all-inclusive approximators, and has been executed in numerous fields. This is because of its capacity to naturally take in any example with no earlier suppositions and loss of all inclusive statement. ANNs have contributed fundamentally towards time arrangement expectation field, yet the nearness of exceptions that normally happen in the time arrangement information may dirty the system preparing information. Hypothetically, the most widely recognized calculation to prepare the system is the backpropagation (BP) calculation which depends on the minimization of the common ordinary least squares (OLS) estimator as far as mean squared error (MSE). Be that as it may, this calculation is not absolutely strong within the sight of exceptions and may bring about the bogus forecast of future qualities. Accordingly, in this paper, we actualize another calculation which exploits firefly calculation on the minimal middle of squares (FA-LMedS) estimator for manufactured neural system nonlinear autoregressive (BPNN-NAR) and counterfeit neural system nonlinear autoregressive moving normal (BPNN-NARMA) models to cook the different degrees of remote issue in time arrangement information. In addition, the execution of the proposed powerful estimator with correlation with the first MSE and strong iterative slightest middle squares (ILMedS) and molecule swarm advancement on minimum middle squares (PSO-LMedS) estimators utilizing reenactment information, in light of root mean squared blunder (RMSE) are likewise talked about in this paper. It was found that the robustified backpropagation learning calculation utilizing FA-LMedS beat the first and other powerful estimators of ILMedS and PSO-LMedS. As a conclusion, developmental calculations beat the first MSE mistake capacity in giving hearty preparing of counterfeit neural systems.
Gravitational Search Algorithm Based Technique for Voltage Stability Improvement Mohamad Khairuzzaman Mohamad Zamani; Ismail Musirin; Mohamad Sabri Omar; Saiful Izwan Suliman; Nor Azura Md. Ghani; Nur Azwan Mohamed Kamari
Indonesian Journal of Electrical Engineering and Computer Science Vol 9, No 1: January 2018
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v9.i1.pp123-130

Abstract

Voltage instability problem has been known as a significant threat to power system operation since its occurrence can lead to power interruption. This phenomenon can be due to uncontrollable load increment, line and generator outage contingencies or unplanned load curtailment. Optimal reactive power dispatch involving reactive power support can be one of the options for improving voltage stability of a power system, which also requires optimization process. Optimal sizing and location can of reactive power support can avoid the system from experiencing over-compensated or under-compensated phenomena. The presence of optimization techniques has helped solving non-optimal phenomenon, nevertheless some setbacks have also been experienced in terms of inaccuracy and stuck in local optima. This paper presents the application of Gravitational Search Algorithm (GSA) technique in attempt to solve optimal reactive power dispatch problem in terms of reactive power support for voltage stability improvement. Optimization process tested on IEEE 14-bus Reliability Test System (RTS) has revealed its superiority with significant promising results in terms of voltage stability improvement in the test system. 
A predictive model for prediction of heart surgery procedure Aufzalina Mohd Yusof; Nor Azura Md. Ghani; Khairul Asri Mohd Ghani; Khairul Izan Mohd Ghani
Indonesian Journal of Electrical Engineering and Computer Science Vol 15, No 3: September 2019
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v15.i3.pp1615-1620

Abstract

Coronary heart disease (CHD) is a disease in which plague in the form of waxy substance builds up inside the coronary arteries. Coronary artery bypass grafting (CABG) is used as treatment on CHD patients but the role of CABG has been challenged by percutaneous coronary intervention (PCI) when it was introduced in 1977.  Drug eluting stents (DES) was introduced with the development of PCI. The purpose of this study was to find the potential risk factors that associated with the procedures (CABG and DES) and to model procedure (CABG vs DES) on coronary heart disease male patients aged 45 years old and below. The study sample was among male patients aged 45 years old and below who has undergone CABG or DES procedure at either IJN or HUKM  from January 2007 until December 2010. Logistic regression was used to model treatment selection on coronary heart disease with 87.3% of the classification rate. Patient who i) smoke, ii) obese, or ii) had dyslipidemia was significantly associated with DES, and the other factors were prone to have CABG as their treatment.