What is ANN?
ANN stand for Artificial Neural Network. Just like a biological neural network (for example your brain), an artificial neural network consist of a population of interconnected neurons that process information, communicates together, learns and adapts to what they learn. They're very useful to model complex relationship between inputs and outputs or to find patterns in data.
The data generated by the stock market is extent and complex, but ANN can help find patterns where humans would have only see randomness.
Resources
If you want to learn more, you can read the Artificial Neural Network page on Wikipedia. I also compiled a list of resources that can be of interest if you want to read about or experiment with neural networks.
Books
- Neural Networks in Finance: Gaining Predictive Edge in the Market
: This book shows how neural networks may be put to work for more accurate forecasting, classification, and dimensionality reduction for better decision making in financial markets.
- Introduction to Neural Networks for Java, Second Edition
: Written by the author of the Encog library, it contains useful examples with source code.
Libraries
- Encog Artificial Intelligence Framework for Java: I discovered this library recently and I have done some preliminary tests with it. So far it performs honestly, but I find it a little bit immature in its present state.
- Fast Artificial Neural Network Library (FANN): This is the library ANNcse currently use. It's fast and easy to use.
- Joone - Java Object Oriented Neural Engine: This is the first library I used, but I finded it too slow for the tasks ANNcse needs to perform. This project hasn't been updated in a while.
Articles / Websites
- Artificial Neural Networks: Short introduction article about ANN. You will find basic principles behind ANN and some typical uses.
- Neural Networks - Training Models and algorithms: Website dedicated to neural networks theories and the mathematics underneath them.
- Neural Network Tutorial: Website dedicated to neural networks and genetic algorithms with detailed tutorials and code examples.