Research on K-Value Selection Method of K-Means Clustering Algorithm

Abstract

Among many clustering algorithms, the K-means clustering algorithm is widely used because of its simple algorithm and fast convergence. However, the K-value of clustering needs to be given in advance and the choice of K-value directly affect the convergence result. To solve this problem, we mainly analyze four K-value selection algorithms, namely Elbow Method, Gap Statistic, Silhouette Coefficient, and Canopy; give the pseudo code of the algorithm; and use the standard data set Iris for experimental verification. Finally, the verification results are evaluated, the advantages and disadvantages of the above four algorithms in a K-value selection are given, and the clustering range of the data set is pointed out.

The Effects of Amelogenin Exon 5 Encoded Peptide from Enamel Matrix Derivative Enhances in Odontoblast like Cells, KN-3 Cells

“…EMD has been reported to induce the production of anti-EMD antibodies in the host [11]; it is generally recognized that only peptides of greater than approximately 10 residues (or > 5 kDa) can function as antigens [12,13]. Thus, because SP is only seven amino acids in length and 1,118 Da in molecular mass, it exhibits very little risk of eliciting an immunological response.…”

Analyze K-Value Selected Method of K-Means Clustering Algorithm to Clustering Province Based on Disease Case

“…The Elbow method is a -value method that determines the best value by looking at the number of places that will form Elbouw at a point [8]. Performance indicators use number of squared errors (SSE).Clusters are said to be convergent when obtaining a smaller value compared to others [2]. SSE formula in K-Means [8]: (5) The Elbow algorithm method in K-Means as follows [9]: 1.…”

“…Input data that has been calculated by Euclidean distance will be one cluster if the distance between the input data with centroid has the smallest value compared to the distance between the input data and centroid on the other cluster. The new centroid of cluster is calculated using the formula: (2) where:…”

“…The number of clusters in the k-means clustering algorithm is determined by finding the best value. Determination of the value of value produces a cluster of data that has been inputted [2]. The identification of determining the number of clusters is the most important at the first step in the K-Means algorithm.…”

“…The most important initial step in the K-Means algorithm is determining the value of the number of clusters or values. The problem in this algorithm is usually the number of clusters does not have the right way to choose the value of k [2]. If the determination of value is not good, it will affect the results of clustering.…”