Jaccard index from a dataframe/array

I am trying to calculate the Jaccard Index of an array/DataFrame using the package Distance.jl, and am having difficulty.

using Distances, DataFrames

zxcv = [0.0 0.0 1.0;0.0 0.0 1.0;0.0 0.0 0.0]
dist = evaluate(jaccard(),zxcv[:,1],zxcv[:,2]) # error no method matching jaccard

In R the formula is

zxcv<-c(0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0)
dim(zxcv)<-c(3,3)

trophic.sim = function(web){
    # compute trophic similarity (Jaccard index)
    # using the cocitation matrix approach
    # cocite defines the cardinal of the intersection of prey and preds
    cocite = web%*%t(web) + t(web)%*%web  # |a inter b|
    # now calculate union as |a| + |b| - |a inter b|
    web.degree = colSums(web) + rowSums(web) # |a| + |b|
    nb_s = nrow(web)
    sum.degree = outer(web.degree,web.degree, FUN = '+') # sum.degree contains |a| + |b|
    web.union = sum.degree - cocite
    obs.trophic.sim = cocite / web.union
    return(obs.trophic.sim)
}
 
mat.similarity = trophic.sim(zxcv)

Could someone show me how to use the Distance package correctly, or what the R code would look like in Julia?

4 posts - 3 participants

Read full topic