Saturday, 15 June 2013

cluster analysis - Clustering by distance in R -


I have the integer vectors that I would like to divide into clusters so that the distance between any two clusters is less than Is bound, and within any cluster, the distance between the two elements is less than the upper limit. For example, suppose we have the following vector:

1, 4, 5, 6, 9, 29, 32, 36

and below 9 and 9 respectively Set the left and upper limits below, the two vectors below should have a possible outcome:

Thanks to Flodell's comments, I realized that this kind of clustering could be impossible. 5, 6, 9

29, 32, 36


So I would like to revise the questions slightly:

If I only between If the cluster reduces the distance, then what are possible clustering methods? If I only within cluster distance bound upper

What are the possible clustering methods?

If the cluster is reduced in the middle of the distance, then what are the possible clustering methods? Hierarchical clustering with

Single Linking : >

  x   

Hierarchical clustering with complete linking : >

  x <-c (1, 4, 5, 6, 9, 20, 26, 29, 32 ) Tree & lt; - Haxtst (dist (x), method = "full") partition (X, caterers (tree, H = 9)) # $ `1` # [1] 1 4 5 6 9 # # $` 2` # [1] 20 # # $ `3` # [1] 26 29 32    

No comments:

Post a Comment