StreamingKMeans (S-Space Package 2.0.1 API)

java.lang.Object
- edu.ucla.sspace.clustering.StreamingKMeans<T>

All Implemented Interfaces:

Generator<OnlineClustering<T>>
```
public class StreamingKMeans<T extends DoubleVector>
extends Object
implements Generator<OnlineClustering<T>>
```
An implementation of a simple, highly accurate streaming K Means algorithm. It is based on a the following paper:
- Braverman, V., Meyerson, A., Ostrovsky, R., Roytman, A., Shindler, M., and Tagiku, B. Streaming k-means on Well-Clusterable Data. In Proceedings of SODA. 2011, 26-40. Available online here
A key feature of this algorithm is that only one pass is made through a data set. It is intended for applications where the total number of data points is known, but can be used if a rough estimate of the data points is known. This algorithm periodically reformulates the centroids by performing an batch form of K Means over the centroids, and potentially a sample of data points assigned to each centroid, clearing out all centroids, and then treating the old centroids as new heavily weighted data points. This process happens automatically when one of several thresholds are passed.
Author:

Keith Stevens

Nested Class Summary

Nested Classes
Modifier and Type Class and Description

class StreamingKMeans.StreamingKMeansClustering<T extends DoubleVector>
The internal OnlineClustering implementation.

Nested Classes
Modifier and Type	Class and Description
`class`	`StreamingKMeans.StreamingKMeansClustering<T extends DoubleVector>` The internal `OnlineClustering` implementation.

Field Summary

Fields
Modifier and Type	Field and Description
`static String`	`ALPHA_PROPERTY` An alpha value, see page 6 in the paper for details.
`static String`	`BETA_PROPERTY` A beta value, see page 6 in the paper for details.
`static String`	`COFL_PROPERTY`
`static double`	`DEFAULT_ALPHA` The default alpha value.
`static double`	`DEFAULT_BETA` The default beta value.
`static double`	`DEFAULT_COFL`
`static double`	`DEFAULT_GAMMA` The default gamma value.
`static double`	`DEFAULT_KOFL`
`static int`	`DEFAULT_NUM_CLUSTERS` The default number of clusters.
`static int`	`DEFAULT_NUM_POINTS` The default number of clusters.
`static String`	`GAMMA_PROPERTY` A gamma value, see page 6 in the paper for details.
`static String`	`KOFL_PROPERTY`
`static String`	`NUM_POINTS_PROPERTY` An estimate of the total number of data points that will be clustered.

Constructor Summary

Constructors
Constructor and Description
`StreamingKMeans()` Creates a new generator using the system properties.
`StreamingKMeans(Properties props)` Creates a new generator using the given properties.

Method Summary

Methods
Modifier and Type	Method and Description
`OnlineClustering<T>`	`generate()` Generates a new instance of a `StreamingClustering` based on the values used to construct this generator.
`String`	`toString()` Returns "StreamingKMeans"

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

- Field Detail
  - NUM_POINTS_PROPERTY
```
public static final String NUM_POINTS_PROPERTY
```
    An estimate of the total number of data points that will be clustered.
    
    See Also:
    Constant Field Values
  - ALPHA_PROPERTY
```
public static final String ALPHA_PROPERTY
```
    An alpha value, see page 6 in the paper for details.
    
    See Also:
    Constant Field Values
  - COFL_PROPERTY
```
public static final String COFL_PROPERTY
```
    See Also:
    Constant Field Values
  - KOFL_PROPERTY
```
public static final String KOFL_PROPERTY
```
    See Also:
    Constant Field Values
  - BETA_PROPERTY
```
public static final String BETA_PROPERTY
```
    A beta value, see page 6 in the paper for details.
    
    See Also:
    Constant Field Values
  - GAMMA_PROPERTY
```
public static final String GAMMA_PROPERTY
```
    A gamma value, see page 6 in the paper for details.
    
    See Also:
    Constant Field Values
  - DEFAULT_NUM_CLUSTERS
```
public static final int DEFAULT_NUM_CLUSTERS
```
    The default number of clusters.
    
    See Also:
    Constant Field Values
  - DEFAULT_NUM_POINTS
```
public static final int DEFAULT_NUM_POINTS
```
    The default number of clusters.
    
    See Also:
    Constant Field Values
  - DEFAULT_ALPHA
```
public static final double DEFAULT_ALPHA
```
    The default alpha value.
    
    See Also:
    Constant Field Values
  - DEFAULT_COFL
```
public static final double DEFAULT_COFL
```
    See Also:
    Constant Field Values
  - DEFAULT_KOFL
```
public static final double DEFAULT_KOFL
```
    See Also:
    Constant Field Values
  - DEFAULT_BETA
```
public static final double DEFAULT_BETA
```
    The default beta value.
    
    See Also:
    Constant Field Values
  - DEFAULT_GAMMA
```
public static final double DEFAULT_GAMMA
```
    The default gamma value.
    
    See Also:
    Constant Field Values
- Constructor Detail
  - StreamingKMeans
```
public StreamingKMeans()
```
    Creates a new generator using the system properties.
  - StreamingKMeans
```
public StreamingKMeans(Properties props)
```
    Creates a new generator using the given properties.
- Method Detail
  - generate
```
public OnlineClustering<T> generate()
```
    Generates a new instance of a StreamingClustering based on the values used to construct this generator.
    
    Specified by:
    
    generate in interface Generator<OnlineClustering<T extends DoubleVector>>
  - toString
```
public String toString()
```
    Returns "StreamingKMeans"
    
    Overrides:
    
    toString in class Object

Class StreamingKMeans<T extends DoubleVector>

Nested Class Summary

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

NUM_POINTS_PROPERTY

ALPHA_PROPERTY

COFL_PROPERTY

KOFL_PROPERTY

BETA_PROPERTY

GAMMA_PROPERTY

DEFAULT_NUM_CLUSTERS

DEFAULT_NUM_POINTS

DEFAULT_ALPHA

DEFAULT_COFL

DEFAULT_KOFL

DEFAULT_BETA

DEFAULT_GAMMA

Constructor Detail

StreamingKMeans

StreamingKMeans

Method Detail

generate

toString