4.6 Conclusion

Component 1: explains the difference between Viena variants.
Component 2: explains the difference between Pavo variants.
Notable variables: Floury, Rubbery, Dark Meat, White Meat, Umami (aka HVP), Spicy, Bitter, and Acidic.

MCA provided similar interpretation as PCA in term of Row Factor Scores and seem to be more selective in term of Column Factor Score. In other words, MCA provided new insights into the columns and their relationships, further explain the correlation between variables that have non-linear relationships. Generally, Component 1 explains the difference between Viena variants and Component 2 explains the difference between Pavo variants. We can also see that binning on a qualitative dataset like the one we have can be challenging while yielding similar results. Please refer to CA method to see how appropriate binning on qualitative (categorical) data can be much easier and more insightful. I also would like to suggest using other clustering methods such as K-means then use both parametric and non-parametric analyses to determing the number of appropriate bins.