Aimfird
commented
1 year ago Hello ! I don't know if you found a solution to your problem yet, but I had quite the same issue today, which I fixed.
Indeed, the groups obtained using the function HCPC were not the same we can see in the fviz_dend dendrogramme I was doing after, but also not the same than the HCPC Dendrogramme (this dendrogramme appears first when running HCPC function, so its hidden by the "Factor map" at the end in the "plots" side of R Studio).
When I saw that even the HCPC dendrogramme was different, I realized the problem was in the HCPC function, and thanks to this post : https://groups.google.com/g/factominer-users/c/cbdTMzx0LVM , I learned that HCPC function was doing a "consolidation" after the dendrogramme (consolidation I do not really understand). To avoid this, you just need to add "consol=FALSE" in HCPC function.
I hope it helps you ! :)
Hello and thank you for this nice package and the visualization options!
Unfortunately, several users seem to report the issue that from one and the same res.hcpc object the cluster dendrogram and the PCA-style plot (you called it 'Factor map' on your website here: HCPC Guide on STHDA) are not matching. With our data it currently seems to be more the rule than the exception in was also mentioned on the website above, but I haven't found any solutions or guidance on it yet.
Thank you in advance!