Non-hierarchical Clustering using k-Means and EM Algorithms

Purpose - The data mining techniques are applied in various fields as a method to extract information based on massive data utilized as useful modeling techniques. However, it is a difficult problem to present the perfect result while considering all circumstances if applied in actual problems, as multiple variables that affects the result exists. Research design, data, methodology - In this paper, non-hierarchical clustering algorithms of k-means and EM are compared through experiments. Clustering is given observations of similar things grouped together. It is characterized by identifying the structure of the entire data analysis method. The entire data into several clusters is able to isolate the technique of maximum purpose. Results - Comparing two algorithms, experimental data are described with a brief introduction to clustering techniques for k-means and EM algorithm for the description. Conclusions - With consideration from the experimental results, experimental method was proposed in the two algorithms for comparison with the results for the experiments