A test for spatial homogeneity in cluster analysis

Dubes, Richard C.; Zeng, Guangzhou

doi:10.1007/bf01890074

Cited by 24 publications

(7 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to Lawson and Jures [44], "if too few points are chosen, then the nearest-neighbor distances chosen will not be representative of the entire distribution of distances." If too many points are chosen, [45] warn that the "assumptions about the Beta distribution will be invalid." Previous authors recommend sampling 5 − 10% of the data [44,46].…”

Section: Clusterability Via Spatial Randomnessmentioning

confidence: 99%

To cluster, or not to cluster: An analysis of clusterability methods

Adolfsson

Ackerman

Brownstein

2019

Pattern Recognition

206

117

View full text Add to dashboard Cite

Clustering is an essential data mining tool that aims to discover inherent cluster structure in data. For most applications, applying clustering is only appropriate when cluster structure is present. As such, the study of clusterability, which evaluates whether data possesses such structure, is an integral part of cluster analysis. However, methods for evaluating clusterability vary radically, making it challenging to select a suitable measure. In this paper, we perform an extensive comparison of measures of clusterability and provide guidelines that clustering users can reference to select suitable measures for their applications.

show abstract

Section: Clusterability Via Spatial Randomnessmentioning

confidence: 99%

To cluster, or not to cluster: An analysis of clusterability methods

Adolfsson

Ackerman

Brownstein

2019

Pattern Recognition

206

117

View full text Add to dashboard Cite

show abstract

“…One of the most important subjects in cluster analysis is to understand the spatial relationships between data objects in each cluster, such as dense or sparse regions in a dataset [10]. It becomes a problem, when some clustering algorithms do not obey these relationships and distributions.…”

Section: Data Distribution Patterns For Clusteringmentioning

confidence: 99%

A new approach to improve the accuracy of online clustering algorithm based on scatter/gather model

Farsandaj

Sadeghian

2010

2010 Annual Meeting of the North American Fuzzy Information Processing Society

View full text Add to dashboard Cite

In cluster analysis process used in data mining which enables extracting interesting data patterns from datasets, accuracy and efficiency are the factors which play a pivotal role. Scatter/Gather is a cluster-based browsing model, and most of previous works on this model focused on efficiency of the clustering algorithm. In this paper we present an algorithm which could improve the accuracy of the online clustering algorithm while still maintain a reasonable level of efficiency. Our experiment proves that the new algorithm is more accurate than the original algorithm.

show abstract

“…U j are the minimum distances of the sub-models from m random points in the sampling window. To define the sampling window, we either took 25 to 75 percentile of the feature values or from δ to max.value-δ along each dimension, where δ denotes the standard deviation of the feature value (Dubes & Zeng 1987; Zeng & Richard C Dubes 1985; Zeng & Richard C. Dubes 1985). To estimate p-value, we repeat the above procedure 1000 times and measured the H value.…”

Section: Methodsmentioning

confidence: 99%

Heterogeneity of Transcription Factor binding specificity models within and across cell lines

Sharmin

Bravo

Hannenhalli

2015

Preprint

View full text Add to dashboard Cite

Complex gene expression patterns are mediated by binding of transcription factors (TF) to specific genomic loci. The in vivo occupancy of a TF is, in large part, determined by the TFs DNA binding interaction partners, motivating genomic context based models of TF occupancy. However, the approaches thus far have assumed a uniform binding model to explain genome wide bound sites for a TF in a cell-type and as such heterogeneity of TF occupancy models, and the extent to which binding rules underlying a TFs occupancy are shared across cell types, has not been investigated. Here, we develop an ensemble based approach (TRISECT) to identify heterogeneous binding rules of cell-type specific TF occupancy and analyze the inter-cell-type sharing of such rules. Comprehensive analysis of 23 TFs, each with ChIP-Seq data in 4-12 cell-types, shows that by explicitly capturing the heterogeneity of binding rules, TRISECT accurately identifies in vivo TF occupancy (93%) substantially improving upon previous methods. Importantly, many of the binding rules derived from individual cell-types are shared across cell-types and reveal distinct yet functionally coherent putative target genes in different cell-types. Closer inspection of the predicted cell-type-specific interaction partners provides insights into context-specific functional landscape of a TF. Together, our novel ensemble-based approach reveals, for the first time, a widespread heterogeneity of binding rules, comprising interaction partners within a cell-type, many of which nevertheless transcend cell-types. Notably, the putative targets of shared binding rules in different cell-types, while distinct, exhibit significant functional coherence.

show abstract

A test for spatial homogeneity in cluster analysis

Abstract: Cluster homogeneity, Spatial uniformity, Cluster validity, Data normalization,

Cited by 24 publications

References 32 publications

To cluster, or not to cluster: An analysis of clusterability methods

To cluster, or not to cluster: An analysis of clusterability methods

A new approach to improve the accuracy of online clustering algorithm based on scatter/gather model

Heterogeneity of Transcription Factor binding specificity models within and across cell lines

Contact Info

Product

Resources

About