TY - JOUR AU - Rosly, Sitti Sufiah Atirah AU - Moktar, Balkiah AU - Mohd Razali, Muhamad Hasbullah PY - 2017/03/30 Y2 - 2024/03/29 TI - Comparison of Clustering Algorithms on Air Quality Substances in Peninsular Malaysia JF - Journal of Computing Research and Innovation JA - JCRINN VL - 2 IS - 1 SE - General Computing DO - UR - https://jcrinn.com/index.php/jcrinn/article/view/28 SP - 36-44 AB - <div><em>Air quality is one of the most popular environmental problems in this globalization era. Air</em></div><div><em>pollution is the poisonous air that comes from car emissions, smog, open burning, chemicals</em></div><div><em>from factories and other particles and gases. This</em></div><div><em>harmful air can give adverse effects to</em></div><div><em>human health and the environment. In order to provide information which areas are better for</em></div><div><em>the residents in Malaysia, cluster analysis is used to determine the areas that can be clustering</em></div><div><em>together based on their a</em></div><div><em>ir quality through several air quality substances. Monthly data from</em></div><div><em>37 monitoring stations in Peninsular Malaysia from the year 2013 to 2015 were used in this</em></div><div><em>study. K</em></div><div><em>-</em></div><div><em>Means (KM) clustering algorithm, Expectation Maximization (EM) clustering</em></div><div><em>algorithm and</em></div><div><em>Density Based (DB) clustering algorithm have been chosen as the techniques to</em></div><div><em>analyze the cluster analysis by utilizing the Waikato Environment for Knowledge Analysis</em></div><div><em>(WEKA) tools. Results show that K</em></div><div><em>-</em></div><div><em>means clustering algorithm is the best method among ot</em></div><div><em>her</em></div><div><em>algorithms due to its simplicity and time taken to build the model. The output of K</em></div><div><em>-</em></div><div><em>means</em></div><div><em>clustering algorithm shows that it can cluster the area into two clusters, namely as cluster 0</em></div><div><em>and cluster 1. Clusters 0 consist of 16 monitoring stations and clu</em></div><div><em>ster 1 consists of 36</em></div><div><em>monitoring stations in Peninsular Malaysia.</em></div> ER -