java.lang.Object
- dev.brachtendorf.jimagehash.datastructures.ClusterResult

public class ClusterResult
extends Object

Since:: 3.0.0
Author:: Kilian

Field Summary

Fields
Modifier and Type	Field	Description
`protected int[]`	`clusterIndex`	Keep track to which cluster a certain points belongs
`protected HashMap<Integer,FuzzyHash>`	`clusters`
`protected HashMap<Integer,List<Integer>>`	`entriesInCluster`	Key the cluster index, value the data index
`protected HashMap<Hash,Integer>`	`entryToDataIndex`
`protected int`	`numberOfClusters`
`protected HashMap<Integer,DoubleSummaryStatistics>`	`stats`	Compute the min max average and mean of each cluster

Constructor Summary

Constructors
Constructor Description

ClusterResult(int[] clusterIndex, Hash[] hashes)

Method Summary

All Methods Instance Methods Concrete Methods
Modifier and Type	Method	Description
`List<Integer>`	`clusterIndexToDataIndex(int clusterIndex)`
`int`	`getBestFitCluster(Hash testHash)`	Return the cluster index whose centeroid is most similar to the supplied hash
`FuzzyHash`	`getCenteroid(int cluster)`	Get the fuzzy hash representing the specified cluster
`List<Hash>`	`getCluster(int cluster)`
`int[]`	`getClusterData()`
`Map<Integer,List<Hash>>`	`getClusters()`
`Map<Integer,Double>`	`getPotentialFits(Hash testHash, double sigma)`	Checks which cluster has a high chance to contain the closest neighbor to this hash
`double`	`getSilhouetteCoef(int cluster)`
`DoubleSummaryStatistics`	`getStats(int cluster)`
`double`	`getSumSquaredError()`
`double`	`getSumSquaredError(int cluster)`
`int`	`indexToCluster(int index)`	Return the cluster index for the data point of the given index
`int`	`lookupClusterIdForKnownHash(Hash testHash)`	Return the cluster index for a hash which was used during the clustering.
`void`	`printInformation(boolean includeSilhouetteCoefficient)`

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

- Field Detail
  - numberOfClusters
```
protected int numberOfClusters
```
  - clusterIndex
```
protected int[] clusterIndex
```
    Keep track to which cluster a certain points belongs
  - stats
```
protected HashMap<Integer,DoubleSummaryStatistics> stats
```
    Compute the min max average and mean of each cluster
  - entryToDataIndex
```
protected HashMap<Hash,Integer> entryToDataIndex
```
  - entriesInCluster
```
protected HashMap<Integer,List<Integer>> entriesInCluster
```
    Key the cluster index, value the data index
  - clusters
```
protected HashMap<Integer,FuzzyHash> clusters
```
- Constructor Detail
  - ClusterResult
```
public ClusterResult(int[] clusterIndex,
                     Hash[] hashes)
```
- Method Detail
  - printInformation
```
public void printInformation(boolean includeSilhouetteCoefficient)
```
  - getClusters
```
public Map<Integer,List<Hash>> getClusters()
```
  - getCluster
```
public List<Hash> getCluster(int cluster)
```
  - getStats
```
public DoubleSummaryStatistics getStats(int cluster)
```
  - getClusterData
```
public int[] getClusterData()
```
  - getSumSquaredError
```
public double getSumSquaredError(int cluster)
```
  - getSumSquaredError
```
public double getSumSquaredError()
```
  - getSilhouetteCoef
```
public double getSilhouetteCoef(int cluster)
```
  - getBestFitCluster
```
public int getBestFitCluster(Hash testHash)
```
    Return the cluster index whose centeroid is most similar to the supplied hash
    
    Parameters:
    
    testHash - the hash to check against the clusters
    
    Returns:
    
    the category (index) of the best matching cluster
  - lookupClusterIdForKnownHash
```
public int lookupClusterIdForKnownHash(Hash testHash)
```
    Return the cluster index for a hash which was used during the clustering.
    This method will return the same value as getBestFitCluster(Hash) but is quicker.
    
    Parameters:
    
    testHash - The hash to check the cluster id for.
    
    Returns:
    
    the index of the cluster this hash belongs to
    
    Throws:
    
    NullPointerException - if the hash wasn't used during clustering.
  - clusterIndexToDataIndex
```
public List<Integer> clusterIndexToDataIndex(int clusterIndex)
```
  - indexToCluster
```
public int indexToCluster(int index)
```
    Return the cluster index for the data point of the given index
    
    Parameters:
    
    index - of the datapoint when the cluster method was called
    
    Returns:
    
    the cluster index
  - getCenteroid
```
public FuzzyHash getCenteroid(int cluster)
```
    Get the fuzzy hash representing the specified cluster
    
    Parameters:
    
    cluster - the cluster index
    
    Returns:
    
    the centeroid of the cluster
  - getPotentialFits
```
public Map<Integer,Double> getPotentialFits(Hash testHash,
                                                  double sigma)
```
    Checks which cluster has a high chance to contain the closest neighbor to this hash
    
    Parameters:
    
    testHash - the hash to check which cluster it belongs to
    
    sigma - a stretch factor indicating how much error from a cluster center to the hash is allowed based on the range of the distances within the cluster. With the original dataset a sigma of 1 would include the best fit cluster with a 100% certainty.
    
    Returns:
    
    the cluster indices a match is most likely

Class ClusterResult

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

numberOfClusters

clusterIndex

stats

entryToDataIndex

entriesInCluster

clusters

Constructor Detail

ClusterResult

Method Detail

printInformation

getClusters

getCluster

getStats

getClusterData

getSumSquaredError

getSumSquaredError

getSilhouetteCoef

getBestFitCluster

lookupClusterIdForKnownHash

clusterIndexToDataIndex

indexToCluster

getCenteroid

getPotentialFits