11.07.2015 Views

Preface to First Edition - lib

Preface to First Edition - lib

Preface to First Edition - lib

SHOW MORE
SHOW LESS
  • No tags were found...

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

ANALYSIS USING R 327R> pottery_single pottery_complete pottery_average layout(matrix(1:3, ncol = 3))R> plot(pottery_single, main = "Single Linkage",+ sub = "", xlab = "")R> plot(pottery_complete, main = "Complete Linkage",+ sub = "", xlab = "")R> plot(pottery_average, main = "Average Linkage",+ sub = "", xlab = "")0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5364142 383912132 4141815 832021 56171925292224343510169434537442332263330271140312871Single LinkageHeight0 2 4 6 8 10 12222324121013343035 252932933162 414 181561719320821 5383936 41423744452711312628434071Complete LinkageHeight0 2 4 6383936 41422529343522241213205 821 171962 414 1815374445302332331016926432711312840371Average LinkageHeightFigure 18.5Hierarchical clustering of pottery data and resulting dendrograms.tition given by applying the kmeans procedure and looking for an ‘elbow’ inthe resulting curve (cf. scree plots in fac<strong>to</strong>r analysis). Such a plot can be constructedin R for the planets data using the code displayed with Figure 18.7(note that since the three variables are on very different scales they first need<strong>to</strong> be standardised in some way – here we use the range of each).Sadly Figure 18.7 gives no completely convincing verdict on the number ofgroups we should consider, but using a little imagination ‘little elbows’ canbe spotted at the three and five group solutions. We can find the number ofplanets in each group usingR> planet_kmeans3 table(planet_kmeans3$cluster)1 2 334 53 14The centres of the clusters for the untransformed data can be computed usinga small convenience function© 2010 by Taylor and Francis Group, LLC

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!