The promise of neural networks

What is the promise of Neural Networks? Just this: they may provide better answers to many "real-world" questions. Problems in industry and business are all too frequently beyond the scope of our present generation of computers. They run into trouble if data is incomplete or contains errors, if a "best guess" is needed, or if it isn't clear how a problem should be tackled. Neural networks are already handling this kind of complex task in areas such as machine vision, time series analysis, robotics, control, cost analysis, and even share price and currency prediction. Neural Networks can learn if they are presented with a range of examples, deduce their own rules for solving problems, and produce valid answers from noisy data. Managers in industry and business, policy-makers, researchers - even undergraduates - will find this book an ideal introduction to Neural Networks. The contents are carefully balanced so as to be accessible to the non-technical reader, while providing detailed information for those with more academic requirements. The EC-funded DEANNA Project (Database for European Artificial Neural Network Activity) has constructed a database of current research, which can be used to judge the success of various applications, in order to assist technology transfer. The results of the DEANNA questionnaire are summarized here.