Cluster of thousands of text documents in java

Question

Is there an effective way to clustering text documents? I was thinking about K-Means, but it seems like it takes too much time. Can someone provide me with an efficient method?

Answer 1

If K-Means really does the job and just seems slow, why not try to do it faster? The method I use is random-pause .

It usually happens that there are many opportunities for speeding up, in code that you would not consider a problem, without changing the underlying algorithm. Here is an example.

Answer 2

, java ?, weka , .