Time filter

Source Type

Bakri M.H.,University Teknikal Malaysia MelakaMelaka | Ali R.,University Technology of MARA | Ismail S.,University Technology of MARA
Advanced Science Letters | Year: 2015

The objective of the study is to testify that the use of residential mortgage-backed securities by Cagamas to finance government staff housing in Malaysia has the ability to increase its net assets value even during global financial crisis 2007–2009. In Malaysia, many firms have been reported involve in Asset Back Securities since 1986s where Cagamas, as its national mortgage agency, is a pioneer. This research examines the determinants of primary market spread and measures financial performance of Cagamas Berhad to represent government housing loans in Malaysia. As a preliminary case study of Cagamas Mortgage-Backed Securities as an issuer the financial performance of its issues of Residential Mortgage Backed Securities is measured by key financial ratios in terms of profitability, capitalization and debt coverage for financial years of 2008–2012, and its key performance indicators as measured by net default and prepayment rates. For leverage effects, the firm shows higher debt coverage and higher earning capacities. Interestingly, its cumulative net default rate and prepayment rate for the percentage of principal balance on the purchase date for both conventional and Islamic securitization show lower than their indicative rates. The findings support past studies on benefits of asset securitization and interestingly verifying that residential Mortgage Backed Securities provides highly rated long-term investment to bank institutions, insurance companies and fund managers and increases its net assets value of securitization. Therefore, Residential Mortgage Backed Securities meets its principal activity for financing government staff housing loans successfully with its net default and prepayment rates show steady improvements on year to year basis. © 2015 American Scientific Publishers. All rights reserved.

Idris S.A.,University Graduate Center | Jafar F.A.,University Teknikal Malaysia MelakaMelaka
Lecture Notes in Electrical Engineering | Year: 2015

This project is focusing on corrosion inspection using image. Inspection which have particularly challenging environmental conditions and characteristics, increase the complexity of the inspection operation. By using software image filter to enhance the image data, it is believe that the object recognition technique will be able to analyse the image data accurately. A few software filters have been identified in this works based on textural feature and colour progression factor that are the characteristic of image corrosion. Therefore, in order to obtain suitable software image filter, neural network is use for optimization. The experiment result shows among those identified image enhancement filters for visual corrosion inspection, Wavelet De-noising gives desirable result in terms of Mean Square Error, Peak Signal to Noise Ratio and Neural Network optimization. © Springer International Publishing Switzerland 2015.

Huat T.C.,University Teknikal Malaysia MelakaMelaka | Manap N.A.,University Teknikal Malaysia MelakaMelaka
Lecture Notes in Electrical Engineering | Year: 2015

A good result of triangulation or known as Three-Dimensional (3D) is depending on the smoothness of the disparity depth map that obtained from the stereo matching algorithms. The smoother the disparity depth map, the better results of triangulation can be achieved. This paper presents the evaluation of the existing stereo matching algorithms in the aspects of the speed of computational on depth map obtained. The stereo matching algorithms that we applied for experimental purpose are basic block matching, sub-pixel accuracy and dynamic programming. The dataset of stereo images that used for the experimental purpose are obtained from Middlebury Stereo Datasets. This research is to provide an idea on choosing the better stereo matching algorithms to work on the disparity depth map for the purpose of 3D triangulation applications, as the good result of 3D triangulation is depending on how smooth is the disparity depth map can be obtained. © Springer International Publishing Switzerland 2015.

Anshar K.,University Teknikal Malaysia MelakaMelaka | Herman N.S.,University Teknikal Malaysia MelakaMelaka
Advanced Science Letters | Year: 2014

There are two new approaches that can be used to develop Smartphone application, which can run on different Smartphones platform, i.e., Hybrid and Titanium approach. Since Smartphone Application Framework provides a browser engine as a class then it allows any native application to utilize it to access any web files such as HTML, CSS, JavaScript, JSON, XML, and etc. stored in Smartphone local system or Server. It is known as Hybrid Application. The second approach, i.e., Titanium, is not a Hybrid application, but this approach still provides an API to develop a Hybrid Application. It brings new approach for Geographic Information Systems (GIS) Application works on different Smartphone Platform and interacts with different GIS data in terms of type, format, and source including Smartphone local system. Application and system architecture as a basic requirement to understand all components and processes involved are discussed to develop such application. © 2014 American Scientific Publishers All rights reserved.

Sutanto D.H.,University Teknikal Malaysia MelakaMelaka | Ghani M.K.A.,University Teknikal Malaysia MelakaMelaka
Advanced Science Letters | Year: 2015

Non-communicable disease (NCDs) is the most epidemic disease and high mortality rate in worldwide likely diabetes mellitus, cardiovascular diseases and cancers. NCDs prediction model have problems such as redundant data, missing data, imbalance dataset and irrelevant attribute. In data mining, feature selection can handle irrelevant attribute. This paper considers finding the optimal feature selection for NCDs prediction model. We comprise 18 feature selection, 4 classification algorithms (Naïve Bayes, Support Vector Machine, Neural Network, and Decision Tree) and used 6 NCDs datasets. The result shows that optimally performed feature selection for NCDs prediction are weight by SVM, W-Uncertainty, W-Chi, and CBWA. © 2015 American Scientific Publishers. All rights reserved.

Discover hidden collaborations