1
$\begingroup$

So, i'm working with fuzzy clustering for Mixed data. Then i want to do Visualization for clustering result. Here is my data

> head(x)
x1 x2        x3       x4
A  C    8.461373 27.62996
B  C   10.962334 27.22474
A  C    9.452127 27.57246
B  D    8.196687 27.29332
A  D    8.961367 26.72793
B  C    8.009029 27.97227

i followed this step https://www.r-bloggers.com/clustering-mixed-data-types-in-r/

gower_dist <- daisy(x,
                 metric = "gower")
                 #type = list(logratio = 1))
tsne_obj <- Rtsne(gower_dist1, dims=2 ,is_distance = TRUE)
tsne_data = data.frame(tsne_obj1$Y, factor(g1$clusters))
colnames(tsne_data1)[3] = "cluster"


ggplot(aes(x = X1, y = X2), data = tsne_data1) +
geom_point(aes(color = cluster))

Based on the website, first step is transformed the data using Gower distance (i guess), than applying R-tsne. So My question is : Is it good using Rtsne for mixed data (as Representative the points)? i have doubt, with gower distance in the first step, its like force your categorical data to be numeric data. but one thing that amazed me, my method always give better result than classic method based on the plot. so this is important for me to know better about this, can i use the plot as a tool to measure the goodness of clustering result? because based on the plot, its not difficult to determine which method is better (by plotting clustering result), i give plot images below, i really impressed with it. Classic Method & My Method

$\endgroup$
  • $\begingroup$ It's not clear to me what you're actually asking, could you clarify? $\endgroup$ – jbowman Mar 27 '18 at 19:19
  • $\begingroup$ First is that okay if used rtsne to representative my data points, as u can see the code .. first step is transformed the data with gower distance, the plot relevant or not to used? Second, I applied tht plot into my method clustering and K-Prototype. And the plots were clearly different, so based on this can i say my method give better clustering result than classic method ( see the pictures).Because its difficult to find validity index to evaluate clustering for mixed data, moreover to compare between two method, Like for K-Prototype u cant just use k-means index validity, right @jbowman $\endgroup$ – Jack shephard Mar 27 '18 at 22:43
1
$\begingroup$

TSNE can be quite misleading, and generate "cluster" structures out of nowhere. See other questions on this problem. I would suggest to also plot the continuous values only, and the categorical values only.

As a baseline comparison, also include the trivial group_by "algorithm", which builds one "cluster" for each unique value of (x1,x2). I would not be surprised if this is A) almost exactly the result you got and B) appears to be very good.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.