In MOOC two, we build upon what we've already learnt in the first half. In the first half, you learnt about Rattle, and R, and some other language. Here we're going to look at simple but powerful data mining techniques. You start with clustering which in my opinion is an underrated subject, even in business analytics. It is a way of grouping similar data together and trying to understand what it tells us. We move on to methods of classification and prediction, again simple rules which work very well like finding similar objects, finding rules from data, using decision tree type of approaches to solve problems and we end up with something exciting which is Recommender Systems, which is trying to recommend people as you're familiar with what else to buy, what people like you have bought, right? How these systems work, we'll learn about that. I also will cover an essential aspect of this; how do you know which model is doing better? How do you improve prediction accuracy? Do you need more data? How do you measure performance. All this, will be done using Rattle. In addition, in a couple of places I would like you to use scripts in R, an R script is a step-by-step way of executing. You don't have to write code, but you just have to run the script in our studio and get the results. So the end of this, I'm sure you'll become experts at least in being able to interpret and being to understand what different commands an R accomplish. Look forward to seeing you in class.