I have medical data set with different types of attributes and I have to do the classification based on attributes: there are no class labels in this data set. Suggest to me which algorithm works better for this dataset.
First of all, your question is not that clear. Next time try adding an example of your dataset.
As you probably know, Machine learning algorithms are described as learning a target function (f) that best maps input variables (X) to an output variable (Y): Y = f(X) .
As @ankit-seth mentioned,
If you want to classify the data you need to have the labels, and then you can use ML algorithms to do that.
I'm guessing you need to cluster(group) your data - try reading about K-Means and go from there ...
good luck :)