Principal component analysis (PCA) of
the texture profile analysis of butter cakes showed
that the two principal components (PCs) explained
total variance of 69% (Figure 4). The first principal
component (PC1 = 40% variance) was
characterized positively by springiness, whereas
cohesiveness, gumminess and adhesiveness were
characterized negatively in PC1. The second
principal component (PC2 = 29% variance) was
characterized positively by hardness and
chewiness. The results from PCA and cluster
analysis indicated that the samples could be
classified into three groups (Figure 4). In Cluster
1, butter cakes made from HMRF were
characterized as having high hardness. In Cluster
2, butter cakes made from Niew Ubon flour and associated mixed flours were grouped together
based on high gumminess and cohesiveness. In
Cluster 3, butter cakes made from RD6 and
associated mixed flours were grouped based on
high springiness, adhesiveness and chewiness.