Couple of Simple questions about Artificial Neural Networks

I've been reading Negnevitsky's 'Artificial Intelligence - A guide to Intelligent Systems'.

I'm trying to get my head around ANN's but one thing is thing is bothering me (I think lack of real-world examples is making it hard for me to understand).

• Is there an output node for every possible outcome? Does this have to be implemented into the NN before training?

• Unsupervised Learning is the means of modifying the weights of a NN without specifying the desired output for any input patterns. Can someone tell me a simple real world application of this?

Thank you, I'd appreciate any help.