Supervised data mining is often more useful in a business context but unsupervised data mining can be very valuable to learn more about your customers, competitors and the general lay of the land.
You are viewing a single comment's thread from:
Supervised data mining is often more useful in a business context but unsupervised data mining can be very valuable to learn more about your customers, competitors and the general lay of the land.
What is the difference between the two in terms of how the mining is structured?
Basically, for supervised mining you need what we call "labelled" data, that is, you need a history of values for the variable you want to predict.
Let's use post rewards as an example. Imagine you want to create a model to predict your post rewards based on a set of variable such as post length, topic, day of the week you posted it and any other atribute you may want to include.
That is a typical supervised mining problem. In this case you will need a dataset with all your posts and the rewards you got on each of them. Then you can train a model to predict the rewards of your future posts.
Fascinating.
This is the type of stuff we need added to #leoai. Keep the explanations coming.
Great example.