I have written an artifical neural network (ANN) implementation for myself (it was fun). I am thinking now about where can I use it.
What are the key areas in the real world, where ANN is being used?
+2
A:
Most often for classifying noisy inputs into fixed categories, like handwritten letters into their equivalent character, spoken voice into phonemes, or noisy sensor readings into a set of fixed values. Usually, the set of categories is small (23 letters, couple of dozen phonemes, etc.)
Others will point out how all these things are better done with specialized algorithms....
Jose M Vidal
2010-09-13 23:49:41
+1
A:
I once wrote an ANN to predict the stock market. It succeeded with about 80% accuracy.
The cue here was to first get hold of a couple of million rows of real stock data. I used this data to train the network and prime it for real data. There were about 8-10 input variables and a single output value that would indicate the predicted value of the stock on the next day.
You could also check out the (ancient) ALVINN network where a car learnt to drive by itself by observing road data when a human driver was behind the wheel.
ANNs are also widely used in bioinformatics.
Pedery
2010-09-14 00:27:38
80% accuracy, eh? You must be rich.
Shaggy Frog
2010-09-14 00:40:26
Just out of curiosity: were these 80% measured on the data that were used to train the network or on a separate set? And please, don't get me wrong, but stating accuracy number without first defining it is absolutely meaningless.
bgbg
2010-09-14 07:30:32
The 80% were measured from using a separate set. However, don't get too happy here. I think my ANN was overly eager to just predict that a stock would go up. Since most stocks do go up, it would be right most of the time. For predicting the stock market, aparently Markov models are better, although I've never attempted to compare the methods.
Pedery
2010-09-14 15:47:02
Using this data of yours, create a one-node decision tree, aka a decision stump, with the answer "Yes" to the question "Will XYZ go up tomorrow?". What kind of accuracy does it achieve? I'm going to guess >50%.
Shaggy Frog
2010-09-15 17:51:58
If you insist on being pedantic, there were about 15 input nodes and a single output node. The output node would predict the stock's value on day n+1. Since one stock's value as raw data isn't really comparable to another stock's value, I felt it would be a good idea to compare it with itself to give an indication of whether it would be a good investment or not.
Pedery
2010-09-20 02:43:37
+2
A:
ANNs are an example of a "learning" system, one that "trains" on input data (in some domain) in order to effectively classify (unseen) data in that domain. They've been used for everything from character recognition to computer games and beyond.
If you're trying to find a domain, pick some topic or field that interests you, and see what kinds of classification problems exist there.
Shaggy Frog
2010-09-14 00:48:40