WOLFRAM

Wolfram Language & System Documentation Center
Wolfram Language Home Page »

ClusterClassify
✖
ClusterClassify

✖

generates a ClassifierFunction[…] by partitioning data into clusters of similar elements.

✖
ClusterClassify[data,n]

generates a ClassifierFunction[…] with n clusters.

Details and Options

Examples

open allclose all

Basic Examples  (3)Summary of the most common use cases

Train the ClassifierFunction on some numerical data:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-0ebdz
Out[1]=1

Use the classifier function to classify a new unlabeled example:

In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-sph195
Out[2]=2

Obtain classification probabilities for this example:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-4zp6ss
Out[3]=3

Classify multiple examples:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-holkya
Out[4]=4

Plot the probabilities for the two different classes in the interval {-5,5}:

In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-ydgxpz
Out[5]=5

Train the ClassifierFunction on some colors by requiring the number of classes to be 5:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-szt4fw
Out[1]=1
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-ym3bco
Out[2]=2

Use the ClassifierFunction on some unlabeled data:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-zy2mue
Out[3]=3

Gather the elements by their class number:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-izazxv
Out[4]=4

Train the ClassifierFunction on some strings:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-e6iypw
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-6wp1yu
Out[2]=2
In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-4h237h
Out[3]=3

Gather the elements by their class number:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-4l3vi1
Out[4]=4

Scope  (11)Survey of the scope of standard use cases

Classify real numbers:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-nozwv1
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-n44cqs
Out[2]=2
In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-4f57j9
Out[3]=3

Classify vectors:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-h33lre
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-xgalcu
Out[2]=2
In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-jkkn6p
Out[3]=3

Classify Boolean vectors:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-grm7nn
Out[1]=1
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-w9ozux
Out[2]=2

Use the classifier to assign clusters to a new Boolean True, False vector:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-r4hx1s
Out[3]=3

Use the classifier to assign clusters to a Boolean 1, 0 vector:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-gpdhgz
Out[4]=4

Look at their probabilities:

In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-7itu0a
Out[5]=5

Classify images:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-pquq9z
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-jhuell
Out[2]=2
In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-pv0vkp
Out[3]=3

Use the classifier to cluster new images:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-e2ti2f
In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-wswzl
Out[5]=5

Classify 3D images:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-635tx1
Out[1]=1
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-d9zq0z
Out[2]=2
In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-cy1gnu
Out[3]=3

Classify colors:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-5nph0v
Out[1]=1
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-073b54
Out[2]=2
In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-sm24z4
Out[3]=3

Classify strings:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-gzyhhy
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-514a6y
Out[2]=2

Use the classifier to cluster new strings:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-9ezsw0
In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-xsnfny
Out[4]=4

Classify heterogeneous data:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-7hanau
Out[1]=1
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-g74ir8
Out[2]=2
In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-9cqla9
Out[3]=3

Classify times:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-22bldm
Out[1]=1
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-eamwze
Out[2]=2

Use the classifier to cluster the data:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-hptkis
Out[3]=3

Classify random reals:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-siuq1r
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-qjolvl
Out[2]=2

Look at the classifier information:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-8jy9w4
Out[3]=3

Get a description for the specific method used:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-3pw6hb
Out[4]=4

Generate random points in the plane and visualize them:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-w79a2z
Out[1]=1

Classify the data:

In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-xltr0q
Out[2]=2

Classify new random points in the place:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-63oxik

Visualize the resulting clustering:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-gu6o77
Out[4]=4

Classify the same test data using IndeterminateThreshold:

In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-7s0lw

Visualize the resulting clustering including the Indeterminate cluster:

In[6]:=6
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-orer4m
Out[6]=6

Options  (10)Common values & functionality for each option

CriterionFunction  (1)

Generate some separated data and visualize it:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-00c176
Out[1]=1

Construct a classifier function using the Automatic CriterionFunction:

In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-ys32xz
Out[2]=2

Construct a classifier function using the Calinski–Harabasz index as CriterionFunction:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-klq9cl
Out[3]=3

Compare the two clusterings of the data:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-ryoy9h
Out[4]=4
In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-4d0yr3
Out[5]=5

FeatureExtractor  (1)

Create a ClassifierFunction from a list of images and classify new examples:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-y4rpu8
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-524nko
Out[2]=2
In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-3y3rjz
Out[3]=3

Create a custom FeatureExtractor to extract features:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-ip44vf
Out[4]=4
In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-sdneu4
Out[5]=5
In[6]:=6
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-uh5nlm
Out[6]=6

FeatureNames  (1)

Generate a classifier function and give a name to each feature:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-48k90i
Out[1]=1

Use the association format to assign cluster to a new example:

In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-f3cwzr
Out[2]=2

The list format can still be used:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-6665m0
Out[3]=3

FeatureTypes  (1)

Generate a classifier function assuming numerical and nominal feature types:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-vup77u
Out[1]=1

Generate a classifier function assuming nominal feature types instead:

In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-74cv6w
Out[2]=2

Compare the result on new examples:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-00dri2
Out[3]=3

Method  (2)

Generate some data using uniform distributions:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-1yt3on
Out[1]=1

Classify the data:

In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-9l2r1q
Out[2]=2

Use Information to obtain a method description:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-l6az2q
Out[3]=3

Look at the clustered data:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-f4ht2o
Out[4]=4

Classify the data using k-means:

In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-l7f9d5
Out[5]=5

Look at the clustered data:

In[6]:=6
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-wtjsy4
Out[6]=6

Generate a large dataset using multinormal distributions and visualize it:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-q74jcq
Out[1]=1

Use ClusterClassify to find clusters by specifying the method to use and look at the AbsoluteTiming:

In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-xeshqx
Out[2]=2

Look at the resulting clustering:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-vjyznx
Out[3]=3

Use ClusterClassify to find clusters without specifying the method to use and look at the AbsoluteTiming:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-34apqm
Out[4]=4

MissingValueSynthesis  (1)

Generate a large dataset using multinormal distributions and visualize it:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-l8qqrk
Out[1]=1

Use ClusterClassify to find clusters:

In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-1o92r9
Out[2]=2

Get the top cluster probabilities for a point with missing data:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-07ipm
Out[3]=3

Set the missing value synthesis to replace each missing variable with its estimated most likely value given known values (which is the default behavior):

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-373liq
Out[4]=4

Replace missing variables with random samples conditioned on known values:

In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-vhbupi
Out[5]=5

Get the distribution of likely clusters for the point by replacing missing variables repeatedly with the random sampling strategy:

In[6]:=6
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-yftwgq
Out[6]=6

PerformanceGoal  (1)

Generate a uniformly distributed dataset and visualize it:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-45i81h
Out[1]=1

Obtain a classifier from this data, with an emphasis on training speed:

In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-467v87
Out[2]=2

Assign clusters to some randomly generated data and look at the AbsoluteTiming:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-iibo83
In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-yzdcxp
Out[4]=4

Obtain a classifier from this data, with an emphasis on the speed:

In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-lcg15e
Out[5]=5

Assign clusters to some randomly generated data and look at the AbsoluteTiming compared to the one above:

In[6]:=6
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-8asw2h
Out[6]=6

Visualize the two clusterings for the test data and note how the setting "TrainingSpeed" gives better results:

In[7]:=7
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-zpj9zx
Out[7]=7
In[8]:=8
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-6mc44v
Out[8]=8

RandomSeeding  (1)

Train several classifiers on random colors:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-zqb4xv
Out[1]=1
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-yh1noh

Compute the classifiers on a new color and observe that the result is always the same:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-6yzdlt
Out[3]=3

Train several classifiers on the same colors by using different values of the RandomSeeding option:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-umfvx6

Compute the classifiers on and observe how the classifier differs:

In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-fk5e85
Out[5]=5

Weights  (1)

Generate some separated data containing outliers:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-sfmfd1
In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-lb5snz
Out[2]=2

Clusterize the data:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-3eop7x
Out[3]=3

Use the classifier function to classify the outlier together with another point:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-cnkb6c
Out[4]=4

Clusterize the data, adding a big weight on the outlier:

In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-r6jn4n
Out[5]=5

Use the classifier function to classify the same points:

In[6]:=6
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-9tdd1
Out[6]=6

Applications  (3)Sample problems that can be solved with this function

Train several classifiers on a small, uniformly distributed dataset:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-n6dayj
In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-jeom79

Divide a triangle into segments by using the classifiers on a large number of uniformly distributed random points:

In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-8rclb9
In[7]:=7
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-584u7j
In[8]:=8
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-2j9ber
Out[8]=8

Generate some normally distributed data:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-0mugs5
Out[1]=1

Clusterize the data without specifying the number of classes:

In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-wwu61c
Out[2]=2
In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-6fq6w3
Out[3]=3

Clusterize the data, specifying the number of classes:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-836l2k
Out[4]=4
In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-4sgm92
Out[5]=5

Find dominant colors in an image:

In[1]:=1
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-x986el

Cluster the data given by the array of pixel values of the image:

In[2]:=2
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-wsmc6f

Use the classifier to assign clusters to each pixel:

In[3]:=3
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-b3uybh

Use the classifier function to find four dominant colors:

In[4]:=4
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-3rz0f5
Out[4]=4

Use the classifier to get binary masks for each dominant color:

In[5]:=5
✖
https://wolfram.com/xid/0jz8jj0ja7kd8g-ymqt0f
Out[5]=5
Wolfram Research (2016), ClusterClassify, Wolfram Language function, https://reference.wolfram.com/language/ref/ClusterClassify.html (updated 2020).
✖
Wolfram Research (2016), ClusterClassify, Wolfram Language function, https://reference.wolfram.com/language/ref/ClusterClassify.html (updated 2020).

Text

Wolfram Research (2016), ClusterClassify, Wolfram Language function, https://reference.wolfram.com/language/ref/ClusterClassify.html (updated 2020).

✖
Wolfram Research (2016), ClusterClassify, Wolfram Language function, https://reference.wolfram.com/language/ref/ClusterClassify.html (updated 2020).

CMS

Wolfram Language. 2016. "ClusterClassify." Wolfram Language & System Documentation Center. Wolfram Research. Last Modified 2020. https://reference.wolfram.com/language/ref/ClusterClassify.html.

✖
Wolfram Language. 2016. "ClusterClassify." Wolfram Language & System Documentation Center. Wolfram Research. Last Modified 2020. https://reference.wolfram.com/language/ref/ClusterClassify.html.

APA

Wolfram Language. (2016). ClusterClassify. Wolfram Language & System Documentation Center. Retrieved from https://reference.wolfram.com/language/ref/ClusterClassify.html

✖
Wolfram Language. (2016). ClusterClassify. Wolfram Language & System Documentation Center. Retrieved from https://reference.wolfram.com/language/ref/ClusterClassify.html

BibTeX

@misc{reference.wolfram_2025_clusterclassify, author="Wolfram Research", title="{ClusterClassify}", year="2020", howpublished="\url{https://reference.wolfram.com/language/ref/ClusterClassify.html}", note=[Accessed: 04-June-2025 ]}

✖
@misc{reference.wolfram_2025_clusterclassify, author="Wolfram Research", title="{ClusterClassify}", year="2020", howpublished="\url{https://reference.wolfram.com/language/ref/ClusterClassify.html}", note=[Accessed: 04-June-2025 ]}

BibLaTeX

@online{reference.wolfram_2025_clusterclassify, organization={Wolfram Research}, title={ClusterClassify}, year={2020}, url={https://reference.wolfram.com/language/ref/ClusterClassify.html}, note=[Accessed: 04-June-2025 ]}

✖
@online{reference.wolfram_2025_clusterclassify, organization={Wolfram Research}, title={ClusterClassify}, year={2020}, url={https://reference.wolfram.com/language/ref/ClusterClassify.html}, note=[Accessed: 04-June-2025 ]}