Understanding Supervised Versus Unsupervised Networks

istock Machine learning is really as simple as an algorithm that combs over any large data set and corresponding events, looking for patterns that allow it to predict the event in real data. That can be heart rhythms with heart attacks, mass email with identified spam (to find the new spam), and genetic combinations with diseases. The clearest examples of those are supervised machine learning, but unsupervised machine learning can find patterns and events that might not even be clear from the outset. The one you choose has very much to do with the characteristics of your particular project. Let’s talk about the trade-offs involved with each type. Supervised Neural Networks First, let’s look at what neural networks are supposed to accomplish.  Supervised networks take data sources (images, rows, play…


Link to Full Article: Understanding Supervised Versus Unsupervised Networks