BFR algorithm - Revision history

Tioaeu8943: clarify "independent dimensions"

2025-05-11T14:51:50Z

clarify "independent dimensions"

← Previous revision		Revision as of 14:51, 11 May 2025
Line 1:		Line 1:
	{{Short description\|Vector clustering algorithms}}		{{Short description\|Vector clustering algorithms}}
	{{more citations needed\|date=May 2018}}		{{more citations needed\|date=May 2018}}
	The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a [[centroid]]. The [[mean]] and [[standard deviation]] for a cluster may differ for different dimensions, but the dimensions must be independent.<ref>{{Cite book\|title=Mining of Massive Datasets\|last=Rajaraman\|first=Anand\|last2=Ullman\|first2=Jeffrey\|last3=Leskovec\|first3=Jure\|publisher=Cambridge University Press\|year=2011\|isbn=1107015359\|location=New York, NY, USA\|pages=257–258}}</ref>		The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a [[centroid]]. The [[mean]] and [[standard deviation]] for a cluster may differ for different dimensions, but the dimensions must be independent.<ref>{{Cite book\|title=Mining of Massive Datasets\|last=Rajaraman\|first=Anand\|last2=Ullman\|first2=Jeffrey\|last3=Leskovec\|first3=Jure\|publisher=Cambridge University Press\|year=2011\|isbn=1107015359\|location=New York, NY, USA\|pages=257–258}}</ref> In other words, the data must take the shape of axis-aligned ellipses.

	==References==		==References==

Tioaeu8943: Adding short description: "Vector clustering algorithms"

2025-05-11T14:46:53Z

Adding short description: "Vector clustering algorithms"

← Previous revision		Revision as of 14:46, 11 May 2025
Line 1:		Line 1:
			{{Short description\|Vector clustering algorithms}}
	{{more citations needed\|date=May 2018}}		{{more citations needed\|date=May 2018}}
	The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a [[centroid]]. The [[mean]] and [[standard deviation]] for a cluster may differ for different dimensions, but the dimensions must be independent.<ref>{{Cite book\|title=Mining of Massive Datasets\|last=Rajaraman\|first=Anand\|last2=Ullman\|first2=Jeffrey\|last3=Leskovec\|first3=Jure\|publisher=Cambridge University Press\|year=2011\|isbn=1107015359\|location=New York, NY, USA\|pages=257–258}}</ref>		The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a [[centroid]]. The [[mean]] and [[standard deviation]] for a cluster may differ for different dimensions, but the dimensions must be independent.<ref>{{Cite book\|title=Mining of Massive Datasets\|last=Rajaraman\|first=Anand\|last2=Ullman\|first2=Jeffrey\|last3=Leskovec\|first3=Jure\|publisher=Cambridge University Press\|year=2011\|isbn=1107015359\|location=New York, NY, USA\|pages=257–258}}</ref>

Derek Andrews: added Category:Cluster analysis algorithms; removed {{uncategorized}} using HotCat

2018-05-20T19:13:13Z

added Category:Cluster analysis algorithms; removed {{uncategorized}} using HotCat

← Previous revision		Revision as of 19:13, 20 May 2018
Line 5:		Line 5:
	{{Reflist}}		{{Reflist}}


	{{Uncategorized\|date=May 2018}}

			[[Category:Cluster analysis algorithms]]

Broccoli and Coffee: Add Reflist, added uncategorised tag using AWB

2018-05-19T01:47:10Z

Add Reflist, added uncategorised tag using AWB

← Previous revision		Revision as of 01:47, 19 May 2018
Line 1:		Line 1:
	{{~~refimprove~~\|date=May 2018}}		{{more citations needed\|date=May 2018}}
	The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a [[~~Centroid\|~~centroid]]. The [[mean]] and [[standard deviation]] for a cluster may differ for different dimensions, but the dimensions must be independent.<ref>{{Cite book\|title=Mining of Massive Datasets\|last=Rajaraman\|first=Anand\|last2=Ullman\|first2=Jeffrey\|last3=Leskovec\|first3=Jure\|publisher=Cambridge University Press\|year=2011\|isbn=1107015359\|location=New York, NY, USA\|pages=~~257-258~~}}</ref>		The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a [[centroid]]. The [[mean]] and [[standard deviation]] for a cluster may differ for different dimensions, but the dimensions must be independent.<ref>{{Cite book\|title=Mining of Massive Datasets\|last=Rajaraman\|first=Anand\|last2=Ullman\|first2=Jeffrey\|last3=Leskovec\|first3=Jure\|publisher=Cambridge University Press\|year=2011\|isbn=1107015359\|location=New York, NY, USA\|pages=257–258}}</ref>

			==References==
			{{Reflist}}

			{{Uncategorized\|date=May 2018}}

Discospinster: Added {{refimprove}} tag to article (TW)

2018-05-18T16:50:58Z

Added {{refimprove}} tag to article (TW)

← Previous revision		Revision as of 16:50, 18 May 2018
Line 1:		Line 1:
			{{refimprove\|date=May 2018}}
	The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a [[Centroid\|centroid]]. The [[mean]] and [[standard deviation]] for a cluster may differ for different dimensions, but the dimensions must be independent.<ref>{{Cite book\|title=Mining of Massive Datasets\|last=Rajaraman\|first=Anand\|last2=Ullman\|first2=Jeffrey\|last3=Leskovec\|first3=Jure\|publisher=Cambridge University Press\|year=2011\|isbn=1107015359\|location=New York, NY, USA\|pages=257-258}}</ref>		The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a [[Centroid\|centroid]]. The [[mean]] and [[standard deviation]] for a cluster may differ for different dimensions, but the dimensions must be independent.<ref>{{Cite book\|title=Mining of Massive Datasets\|last=Rajaraman\|first=Anand\|last2=Ullman\|first2=Jeffrey\|last3=Leskovec\|first3=Jure\|publisher=Cambridge University Press\|year=2011\|isbn=1107015359\|location=New York, NY, USA\|pages=257-258}}</ref>

Nbro at 16:48, 18 May 2018

2018-05-18T16:48:46Z

← Previous revision		Revision as of 16:48, 18 May 2018
Line 1:		Line 1:
	The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a centroid. The mean and standard deviation for a cluster may differ for different dimensions, but the dimensions must be independent.<ref>{{Cite book\|title=Mining of Massive Datasets\|last=Rajaraman\|first=Anand\|last2=Ullman\|first2=Jeffrey\|last3=Leskovec\|first3=Jure\|publisher=Cambridge University Press\|year=2011\|isbn=1107015359\|location=New York, NY, USA\|pages=257-258}}</ref>		The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a [[Centroid\|centroid]]. The [[mean]] and [[standard deviation]] for a cluster may differ for different dimensions, but the dimensions must be independent.<ref>{{Cite book\|title=Mining of Massive Datasets\|last=Rajaraman\|first=Anand\|last2=Ullman\|first2=Jeffrey\|last3=Leskovec\|first3=Jure\|publisher=Cambridge University Press\|year=2011\|isbn=1107015359\|location=New York, NY, USA\|pages=257-258}}</ref>

Nbro: Citation of the book from which the text of this article was taken added

2018-05-18T16:47:37Z

Citation of the book from which the text of this article was taken added

← Previous revision		Revision as of 16:47, 18 May 2018
Line 1:		Line 1:
	The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a centroid. The mean and standard deviation for a cluster may differ for different dimensions, but the dimensions must be independent.		The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a centroid. The mean and standard deviation for a cluster may differ for different dimensions, but the dimensions must be independent.<ref>{{Cite book\|title=Mining of Massive Datasets\|last=Rajaraman\|first=Anand\|last2=Ullman\|first2=Jeffrey\|last3=Leskovec\|first3=Jure\|publisher=Cambridge University Press\|year=2011\|isbn=1107015359\|location=New York, NY, USA\|pages=257-258}}</ref>

Nbro at 16:43, 18 May 2018

2018-05-18T16:43:25Z

← Previous revision		Revision as of 16:43, 18 May 2018
Line 1:		Line 1:
	The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a centroid. The mean and standard deviation for a cluster may differ for different dimensions, but the dimensions must be independent.		The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a centroid. The mean and standard deviation for a cluster may differ for different dimensions, but the dimensions must be independent.

Nbro at 16:43, 18 May 2018

2018-05-18T16:43:09Z

← Previous revision		Revision as of 16:43, 18 May 2018
Line 1:		Line 1:
	The '''BFR algorithm''', named after its inventors ��Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a centroid. The mean and standard deviation for a cluster may differ for different dimensions, but the dimensions must be independent.		The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a centroid. The mean and standard deviation for a cluster may differ for different dimensions, but the dimensions must be independent.

Nbro: Introduction to the algorithm added from the book "Mining of massive datasets" (by Rajaraman, Anand and Ullman, Jeffrey David).

2018-05-18T16:42:52Z

Introduction to the algorithm added from the book "Mining of massive datasets" (by Rajaraman, Anand and Ullman, Jeffrey David).

New page

The '''BFR algorithm''', named after its inventors ��Bradley, Fayyad and Reina, is a variant of [[k-means clustering|k-means]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution|normally distributed]] about a centroid. The mean and standard deviation for a cluster may differ for different dimensions, but the dimensions must be independent.

← Previous revision		Revision as of 14:46, 11 May 2025
Line 1:		Line 1:
			{{Short description\|Vector clustering algorithms}}
	{{more citations needed\|date=May 2018}}		{{more citations needed\|date=May 2018}}
	The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a [[centroid]]. The [[mean]] and [[standard deviation]] for a cluster may differ for different dimensions, but the dimensions must be independent.<ref>{{Cite book\|title=Mining of Massive Datasets\|last=Rajaraman\|first=Anand\|last2=Ullman\|first2=Jeffrey\|last3=Leskovec\|first3=Jure\|publisher=Cambridge University Press\|year=2011\|isbn=1107015359\|location=New York, NY, USA\|pages=257–258}}</ref>		The '''BFR algorithm''', named after its inventors Bradley, Fayyad and Reina, is a variant of [[k-means clustering\|k-means algorithm]] that is designed to cluster data in a high-dimensional [[Euclidean space]]. It makes a very strong assumption about the shape of clusters: they must be [[Normal distribution\|normally distributed]] about a [[centroid]]. The [[mean]] and [[standard deviation]] for a cluster may differ for different dimensions, but the dimensions must be independent.<ref>{{Cite book\|title=Mining of Massive Datasets\|last=Rajaraman\|first=Anand\|last2=Ullman\|first2=Jeffrey\|last3=Leskovec\|first3=Jure\|publisher=Cambridge University Press\|year=2011\|isbn=1107015359\|location=New York, NY, USA\|pages=257–258}}</ref>