# cluster.dist: Clustering a Sparse Symmetric Distance Matrix In cba: Clustering for Business Analytics

## Description

Compute a clustering on a sparse symmetric distance matrix using graph cutting.

## Usage

 `1` ```cluster.dist(x, beta) ```

## Arguments

 `x` an object of class `dist`. `beta` the distance threshold.

## Details

This function computes a clustering on an object of class `dist` by cutting the graph induced by the threshold `beta` into all disconnected subgraphs (the clusters). Two nodes are connected by a link if their distance is below the specified threshold. Note that the threshold is not strict, i.e. `>=`.

Note that distances of value `NA` and `NaN` are ignored. This is not strictly correct but avoids computing 2^k possible solutions if there are k `NA` values.

The time complexity is O(n^2) with n the number of rows/columns.

## Value

A factor of cluster labels (indexed 1,2,...,k).

## Note

Fixme: can the time complexity be improved?

Christian Buchta

## See Also

`dist` and `sdists` for distance computation.

## Examples

 ```1 2 3 4 5 6 7 8 9``` ```## 3 clusters (1 = connected) x <- matrix(c(1,1,0,0,0,0, 1,1,0,0,0,0, 0,0,1,1,0,0, 0,0,1,1,0,0, 0,0,0,0,1,1, 0,0,0,0,1,1), ncol=6) c <- cluster.dist(as.dist(!x), beta = 0) # invert and note that 0 >= 0 c ```

### Example output

```Loading required package: grid
Loading required package: proxy

Attaching package: 'proxy'

The following objects are masked from 'package:stats':

as.dist, dist

The following object is masked from 'package:base':

as.matrix

 1 1 2 2 3 3
Levels: 1 2 3
```

cba documentation built on May 2, 2019, 1:39 p.m.