I wrote an article titled “Classification using Perceptrons” that appears in the April 2013 issue of Visual Studio Magazine. See http://visualstudiomagazine.com/articles/2013/04/01/classification-using-perceptrons.aspx. A perceptron is code that models a single biological neuron. Perceptrons were the predecessor to neural networks — a neural network is a collection of interconnected perceptrons.
Although quite limited, perceptrons can be used to perform certain types of machine learning classification. Perceptron classification uses data to construct a perceptron that can place data into one of several categories, for example, determining whether a hospital patient has (+1) or does not have (-1) a disease based on medical test data.