6. Cluster from CA coordinates (three factors ncp=3). Words
Hierarchical Agglomerative Clustering without Contiguity-Constrained
res.HCca <- LexHCca(resLexCA, cluster.CA="words", nb.clust=5, graph=FALSE)
plot(res.HCca)
## Warning: ggrepel: 1026 unlabeled data points (too many overlaps). Consider increasing max.overlaps
To avoid the problem:
“Warning: ggrepel: 1026 unlabeled data points (too many overlaps). Consider increasing max.overlaps”
plot(res.HCca, plot=c("labels"), labels=c(max.overlaps=Inf))
Labels without reppel
plot(res.HCca, plot=c("labels"), labels=c(size=4, force=0))
50 words with highest coordinates
plot(res.HCca, plot="labels", selInd="coord 50", labels=c(max.overlaps=Inf))
Selecting words
plot(res.HCca, plot="labels", selInd=c("España","españoles", "españolas"), labels=(force=0))
Selecting one cluster
plot(res.HCca, plot="labels", selClust=c(1), labels=c(force=0), xlim=c(-1.2,0), ylim=c(-1.2,.2))
Using max.overlaps:
plot(res.HCca, plot="labels", selClust=c(1), labels=c(max.overlaps=Inf), xlim=c(-1.2,0), ylim=c(-1.2,.2))
plot(res.HCca, plot="labels", selClust=c(1), selInd="contrib .2", labels=c(force=0.3, max.overlaps=Inf), xlim=c(-1,0), ylim=c(-1.2,0))
