Clustering and classification: when to use which technique?

Chris Swart

This talk will describe the differences between supervised and unsupervised techniques and the pros and cons of each. It will explain a couple of document classification techniques , such as SVM, decision tree/forest,  and a couple of clustering techniques, such as graph-based ones and Latent Dirichlet Allocation which is a probabilistic graphical model.

We will ask how you evaluate how well is your algorithm doing, and reinforce what kind of problems require which solution and when is it advantageous to combine the two.

Track 1
Location: Date: October 15, 2018 Time: - Chris Swart Chris Swart, Revolver