Research Article
Modification K-Means Model With Local Deviation Method To Improve The Accuracy In Forming Clusters
@INPROCEEDINGS{10.4108/eai.18-7-2019.2288601, author={M Martiano and Muhammad Zarlis and Sutarman Wage}, title={Modification K-Means Model With Local Deviation Method To Improve The Accuracy In Forming Clusters}, proceedings={Proceedings of The 2nd International Conference On Advance And Scientific Innovation, ICASI 2019, 18 July, Banda Aceh, Indonesia}, publisher={EAI}, proceedings_a={ICASI}, year={2019}, month={11}, keywords={k-means clustering deviation and mse}, doi={10.4108/eai.18-7-2019.2288601} }- M Martiano
Muhammad Zarlis
Sutarman Wage
Year: 2019
Modification K-Means Model With Local Deviation Method To Improve The Accuracy In Forming Clusters
ICASI
EAI
DOI: 10.4108/eai.18-7-2019.2288601
Abstract
K-Means is one method in data mining that can be used to clustering data (Winda, 2015). But in the process of clustering K-Means Trying to minimize the number of Euclidean distance from the mean (Ismkhan, 2018) and it depends on the selection of cluster starting point (Han & Kamber, 2012). In this paper the authors try to use local deviation in calculating the cluster center based on distance variables are calculated and determined the mean so that will form the result between Σx and Σy. The results obtained from the modification algorithm is able to reduce the level of MSE (Means Square Error) performed on tests 1 and 2 that have a value of 290.95, while at K-Means MSE levels change in test 1 reaches 508.54 and test 2 reached 881.13 which gave MSE results higher than K-Means.
