A kernel method : in modelling voter's preference in north region / Nor Zakiah Mahmud

The Geographical Information System theory for the Political Geography has dominated more in political science in particular western countries. However in Malaysia, the theory is still new and not widely used in studying the election or voter’s preference and previously, most studies on elections in the political science field did not take into the aspects of GIS's analysis. This project aims to predict the political geography and election using Kernel Method by modelling of voter’s preference. A kernel method is used for predict the outcomes in this study such as from kernel model, it will know the pattern of political geography mapping and voter’s preference. Thus, analyse the election by kernel method of spatial analyst toward voter’s preference will be able to estimate area that will get affected in the election. The data is supported by the general election information for model the result. The data will be used to mapping the DUN. Then, it will classify and evaluate the kernel method. The process of examining the location, attributes and relationship of features in spatial data through overlay and other analytical technique in order to address a question or gain useful knowledge. In addition, the suggested is to provide meaningful presentation of political status to numerous parties, government and nongovernment that may concern about the political growth and election impact. This research can be reference and academic review about the political geography that include the capability of GIS for analysis making. In term of analysis, GIS can provide meaningful and various analysis including pattern and also statistical view from the particular election result. Hence the usefulness of GIS implementation can be spreading to numerous people either to those who are not familiar with GIS and only has basic knowledge about it.