Hands-On Data Science with Anaconda

Implementation via Julia

The first example uses the familiar dataset called iris again. Using the kmeans() function, the program tries to group these plants:

using Gadfly 
using RDatasets 
using Clustering 
iris = dataset("datasets", "iris") 
head(iris) 
features=permutedims(convert(Array, iris[:,1:4]),[2, 1]) 
result=kmeans(features,3)                           
nameX="PetalLength" 
nameY="PetalWidth" 
assignments=result.assignments   
plot(iris, x=nameX,y=nameY,color=assignments,Geom.point) 

The related output is shown here:

For the next example, we try to sort a set of random numbers into 20 clusters. The code is shown here:

using Clustering srand(1234) nRow=5 nCol=1000 x = ...