Neural Networks
Systems designed to tease out meaningful patterns from vast amounts of
data that humans would find difficult to analyze without computer support
How it works:
Program given set of data
Program analyzes data, works out correlations, selects variables to
create patterns
Pattern used to predict outcomes, then results compared to known
results
Program changes pattern by adjusting variable weights or variables
themselves
Repeats process over and over to adjust pattern
When no further adjustment identified, ready to be used to make
predictions for future cases