Getting non-overlapping sets under the same cluster

[ale0xb]

Hello,

I've been playing around with HierarchicalSets and some toy data (see attachment).

The data is loaded from a JSON file and fed into format_sets() like this:

data <- fromJSON(file = "data.json")

universe <- unique(unlist(data))

x <- lapply(data, function(set) {
  universe %in% set
})

setNames = names(x)

colnames(sets) <- setNames
rownames(sets) <- universe

formatted_set <- format_sets(sets)
dataSet <- create_hierarchy(formatted_set)

Then I check the cluster results with: cluster_members(keySet)

Looking at the output, I see that the algorithm clusters together "el-1272" and "el-2132". However, they do not have any elements in common.

Is this expected behavior? If I understood correctly, this should not happen as the homogeneity between these two sets will always be 0.

Can anyone help me to find out what's going on? Thanks

data.json.zip