Synthetic
Cluster Analysis
Datasets used in the paper “A novel optimization approach towards improving separability of clusters“
Datasets used in the paper “Robust clustering algorithm: The use of soft trimming approach”
Datasets used in the paper “Absolute indices for determining compactness, separability and number of clusters”
Fig. 3: Compactness indices of clusters in the synthetic data
Fig. 7: Illustration of an adjacent set using a synthetic data
Fig. 8: Synthetic data sets with 4 clusters and different separability
Semi-Supervised Clustering (SSC)
Must-link (ML) and cannot-link (CL) pairs: sample data generated randomly using Gaussian distribution: there are 70 data points with six ML and six CL pairs. “Data_SSC_points” has the data points, and “Data_SSC_ML” and “Data_SSC_CL” present the location of data points with ML and CL pairs, respectively. The figure below illustrates the data set: the data points are grouped in four clusters where points in each cluster are denoted using the same color: orange for the points in the first cluster; green for the points in the second cluster; purple for the points in the third cluster, and black for the points in the fourth cluster. The points with ML pairs are joined using the blue solid lines, and those with CL pairs are presented using red dashed lines.
Dataset used in the paper “Nonsmooth Optimization-Based Model and Algorithm for Semisupervised Clustering“
