Clustering Multidimensional Spatial Datasets With DBSCAN, OPTICS, BIRCH, K-Means, and Two-Step Methods: A comparative Evaluation of Five Algorithms - Tapa blanda

Sinopsis

This book presents a systematic approach to density estimation and clustering of multidimensional real-life spatial datasets, utilizing density-based clustering methods, DBSCAN, and OPTICS, and compares their clustering performance to that of the traditional centroid-based K-means, the hierarchical BIRCH, and the hybrid two-step clustering algorithms, evaluating the quality of clusters generated by the five clustering approaches through five quality validation indices including the DBCV validation index. The dbscan R package, used for clustering with DBSCAN and OPTICS algorithms, BIRCH within the stream R package, used to efficiently cluster and identify densely populated regions within datasets, supported by the dimension reduction techniques t-SNE using the tsne R package, and principal component analysis through factor analysis in SPSS, offer a robust platform of cluster analysis. This book will be particularly beneficial to those wishing to employ these density-based techniques in research or applications across statistics, data mining and analysis, clinical research, social science, market segmentation, consumer analysis, and many other disciplines.

"Sinopsis" puede pertenecer a otra edición de este libro.