Project (almost) Finished
I’ve been working on my final paper for 6 months + now and it’s come to the finish line. It’s not what I had hoped it would be but it’s still good enough. Instead of creating a new neural network system (which is a long shot in the first place anyway), I’ve managed a slightly modified radial basis function system to compete with a widely adopted MLP for predicting stock market price. I’m taking RBF as a pattern recognition tools because I’m looking on a wide array of data.
Let’s see if the new system can be better than an MLP system ;-)
After I graduate, I still wanted to learn more about this neural network system because it offers us a wide variety of method to solve a wide variety of problems. People said that this kind of method is a ‘black-box’ method because we can’t see how it works. I say they’re wrong because if we design the system, we can clearly see how it works (to some degree). Next stop should be me learning C++ or Java. Which one do you think should I take?
Not C++ nor Java.
Consider Haskell, Erlang, or Ruby.