New answers tagged clustering
0
votes
How to choose a fair gamma value when performing k-prototypes clustering?
Note that when the numerical features are scaled (z = (x-u)/s, where u is the mean and s the standard deviation), the default gamma value that is calculated will be 0.5 which likely will give very low ...
- 1
1
vote
What's the purity of random clustering?
I don't think assigning points randomly to clusters is a good null model -- I think you'd want something more like clustering data that didn't have clusters in some sense. However.
You don't specify ...
- 43.8k
1
vote
Accepted
Stochastic blockmodel with continuous weights in [0,1]
You are right that the gaussian distribution refers to the numerical weights of the edges, and you are right that the marginal distribution of these edges is not where the gaussian is fitted, but ...
- 4,620
Top 50 recent answers are included