(Automatic) Cluster Count Extraction from Unlabeled Data Sets

Sledge, Isaac J.; Huband, Jacalyn M.; Bezdek, James C.

doi:10.1109/fskd.2008.552

Cited by 20 publications

(13 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After that we apply 2D FFT, IFFT, and correlation on getting VAT image. These steps are described by the authors of [15], and then we obtained histogram in Fig 8. Finally, the clustering number is extracted; it is referred as clustering tendency. The running is reduced in new method when compare of existing system.…”

Section: Results Analysismentioning

confidence: 99%

Reducing Runtime Values in Minimum Spanning Tree Based Clustering by Visual Access Tendency

Reddy¹,

Prasad²

2012

IJDKP

View full text Add to dashboard Cite

Clustering has been widely used in data analysis. Dissimilarity assesses the distance between objects and this is important in Minimum Spanning Tree (MST) based clustering. An inconsistent edge is identified and removed without knowledge of prior tendency in MST based clustering, which explore the results of clusters in the form of sub-trees. Clustering validity is to be checked at every iterated MST clusters by

show abstract

Section: Results Analysismentioning

confidence: 99%

Reducing Runtime Values in Minimum Spanning Tree Based Clustering by Visual Access Tendency

Reddy¹,

Prasad²

2012

IJDKP

View full text Add to dashboard Cite

show abstract

“…Consideration of these two issues is nearly a paper unto itself and would take us far afield from our present objective; hence, this question is taken up in Refs. 34 36 and DBE 37 algorithms extract the number of apparent clusters from VAT images using similar image-processing approaches that differ mainly in the details of the image processing itself. But these two methods stop short of answering the last question.…”

Section: Meanmentioning

confidence: 99%

Clustering in ordered dissimilarity data

Havens

Bezdek

Keller

et al. 2009

Int. J. Intell. Syst.

Self Cite

View full text Add to dashboard Cite

This paper presents a new technique for clustering either object or relational data. First, the data are represented as a matrix D of dissimilarity values. D is reordered to D * using a visual assessment of cluster tendency algorithm. If the data contain clusters, they are suggested by visually apparent dark squares arrayed along the main diagonal of an image I (D * ) of D * . The suggested clusters in the object set underlying the reordered relational data are found by defining an objective function that recognizes this blocky structure in the reordered data. The objective function is optimized when the boundaries in I (D * ) are matched by those in an aligned partition of the objects. The objective function combines measures of contrast and edginess and is optimized by particle swarm optimization. We prove that the set of aligned partitions is exponentially smaller than the set of partitions that needs to be searched if clusters are sought in D. Six numerical examples are given to illustrate various facets of the algorithm. C 2009 Wiley Periodicals, Inc.

show abstract

“…In the following sections, the performance of E-DBE is compared with other preclustering assessment of cluster tendency techniques like DBE [25] and CCE algorithm [12]. CCE also counts dark blocks in RDIs using image transformation techniques.…”

Section: Cluster Count Extraction (Cce) For Cluster Tendency Performancementioning

confidence: 99%

“…In this algorithm VAT [9] was used to obtain RDIs from D, but E-VAT [11] is used in the proposed E-DBE algorithm. CCE algorithm is performed based on the parameter settings suggested in [12], i.e., s′ = 20, p = 1 and w = 0. Section 7 analyzes the results of CCE with DBE and proposed E-DBE on various synthetic, UCI Repository and Real-world datasets.…”

Section: Outputmentioning

confidence: 99%

See 1 more Smart Citation

Enhanced Dark Block Extraction Method Performed Automatically to Determine the Number of Clusters in Unlabeled Data Sets

Prabhu

Duraiswamy

2013

INT J COMPUT COMMUN

View full text Add to dashboard Cite

One of the major issues in data cluster analysis is to decide the number of clusters or groups from a set of unlabeled data. In addition, the presentation of cluster should be analyzed to provide the accuracy of clustering objects. This paper propose a new method called Enhanced-Dark Block Extraction (E-DBE), which automatically identifies the number of objects groups in unlabeled datasets. The proposed algorithm relies on the available algorithm for visual assessment of cluster tendency of a dataset, by using several common signal and image processing techniques. The method includes the following steps: 1.Generating an Enhanced Visual Assessment Tendency (E-VAT) image from a dissimilarity matrix which is the input for E-DBE algorithm. 2. Processing image segmentation on E-VAT image to obtain a binary image then performs filter techniques. 3. Performing distance transformation to the filtered binary image and projecting the pixels in the main diagonal alignment of the image to figure a projection signal. 4. Smoothing the outcrop signal, computing its first-order derivative and then detecting major peaks and valleys in the resulting signal to acquire the number of clusters. E-DBE is a parameter-free algorithm to perform cluster analysis. Experiments of the method are presented on several UCI, synthetic and real world datasets.

show abstract

(Automatic) Cluster Count Extraction from Unlabeled Data Sets

Cited by 20 publications

References 16 publications

Reducing Runtime Values in Minimum Spanning Tree Based Clustering by Visual Access Tendency

Reducing Runtime Values in Minimum Spanning Tree Based Clustering by Visual Access Tendency

Clustering in ordered dissimilarity data

Enhanced Dark Block Extraction Method Performed Automatically to Determine the Number of Clusters in Unlabeled Data Sets

Contact Info

Product

Resources

About