Single-link characteristics of a mode-seeking clustering algorithm

Shaffer, Edward; Dubes, Richard C.; Jain, Anil K.

doi:10.1016/0031-3203(79)90030-x

Cited by 16 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…His choice of MST was influenced by the Gestalt principle, which favors the grouping of attribute patterns based on Euclidean distance measure. Shaffer et al [21] demonstrate the similarity of the mode-seeking partitioning algorithm [12] to the graph algorithm of Zahn [23] based on minimum spanning trees. Lu and Fu [13] used another graph-based approach called "Nearest-Neighbor clustering algorithm" to cluster patterns during character recognition [11].…”

Section: Related Workmentioning

confidence: 98%

An objective function for vertically partitioning relations in distributed databases and its analysis

Chakravarthy

Muthuraj

Varadarajan

et al. 1994

Distrib Parallel Databases

View full text Add to dashboard Cite

Abstract. Partitioning and allocation of relations is an important component of the distributed database design. Several approaches (and algorithms) have been proposed for clustering data for pattern classification and for partitioning relations in distributed databases. Most of the approaches used for classification use square-error criterion. In contrast, most of the approaches proposed for partitioning of relations are either ad hoc solutions or solutions for special cases (e.g., binary vc.r tical partitioning).In this paper, we first highlight the differences between the approaches taken for pattern classification and for distributed databases. Then an objective function for vertical partitioning of relations is derived using the square-error criterion commonly used in data clustering. The objective function derived generalizes and subsumes earlier work on vertical partitioning. Furthermore, the approach proposed in this paper is shown to be useful for comparing previously developed algorithms for vertical partitioning. The objective function has also been extended to include additional information, such as transaction types, different local and remote accessing costs and replication. Finally, we discuss the implementation of a distributed database design testbed.

show abstract

Section: Related Workmentioning

confidence: 98%

An objective function for vertically partitioning relations in distributed databases and its analysis

Chakravarthy

Muthuraj

Varadarajan

et al. 1994

Distrib Parallel Databases

View full text Add to dashboard Cite

show abstract

“…Subsequently, we can analyze the SD of the distance within each tumor type as shown in Figure 6 A. Furthermore, for intercluster analysis, centroids from each tumor type are used to calculate the single linkage among tumor types 52 , 53 , 54 ( Fig. 6 B).…”

Section: Resultsmentioning

confidence: 99%

A Preliminary Investigation into Search and Matching for Tumor Discrimination in World Health Organization Breast Taxonomy Using Deep Networks

Shafique,

Gonzalez,

Pantanowitz

et al. 2024

Modern Pathology

View full text Add to dashboard Cite

“…kNN mode seeking is originally described by Koontz [9]. It is related to an algorithm studied by Kittler [8] for which Shaffer et al [10] stated that although it is based on another idea, its results may be very similar to single-linkage hierarchical clustering. Our experiments have shown that this is not true for our version.…”

Section: The Algorithmsmentioning

confidence: 99%

Mode Seeking Clustering by KNN and Mean Shift Evaluated

Duin

Fred²,

Loog

et al. 2012

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Clustering by mode seeking is most popular using the mean shift algorithm. A less well known alternative with different properties on the computational complexity is kNN mode seeking, based on the nearest neighbor rule instead of the Parzen kernel density estimator. It is faster and allows for much higher dimensionalities. We compare the performances of both procedures using a number of labeled datasets. The retrieved clusters are compared with the given class labels. In addition, the properties of the procedures are investigated for prototype selection. It is shown that kNN mode seeking is well performing and is feasible for large scale problems with hundreds of dimensions and up to a hundred thousand data points. The mean shift algorithm may perform better than kNN mode seeking for smaller dataset sizes.

show abstract

Single-link characteristics of a mode-seeking clustering algorithm

Cited by 16 publications

References 9 publications

An objective function for vertically partitioning relations in distributed databases and its analysis

An objective function for vertically partitioning relations in distributed databases and its analysis

A Preliminary Investigation into Search and Matching for Tumor Discrimination in World Health Organization Breast Taxonomy Using Deep Networks

Mode Seeking Clustering by KNN and Mean Shift Evaluated

Contact Info

Product

Resources

About