CZ4034 2015-2015 Semester 2 Question 3a
Closed this issue · 4 comments
k-means can be used for supervised learning too. I will use both clustering algo, and evaluate using the measures such as precision, recall and f-measure. Should probably use the classifier with a high f1-measure.
@kklw Can you explain in details how k-means clustering can be used to supervised learning? or can you give any references to this?
And since you mentioned use both clustering algo, do you think that kNN is a clustering algorithm?
Sorry, I made a mistake for this. Agree that knn is more suitable for classification.
My impression was i read that kmeans can be used for classification too, but they are modified k-means. E.g. Adapting k-means for supervised
this paper demonstrates how the popular k-means clustering algorithm can be profitably modified to be used as a classifier algorithm.