Question: Why are neural networks better at classification problems then conventional computing systems?
Answer: Neural networks work by simulating what the brain does in working on tasks. The NN works in parallel whilst traditional computing programs work serially with one task following the other.
… most important of the features is that the brain is able to learn things – it can teach itself. Learning from example is the way in which as children we pick up speech, learn to write …
(R Beale & T Jackson) Neural Computing an introduction , page 4
The same is not true of conventional computer systems. In these the computer usually has a long and complicated program, which gives it specific instructions as to what to do at every stage in the operation