Resumen
The article focuses on neural networks. With advancing computer technology, automated knowledge discovery has become an important. All research topic, as well as a practical business application in an increasing number of organizations. Knowledge discovery can be defined as the learning of implicit and previously unknown nontrivial knowledge from data or observations. The intelligence emerging from interactions among numerous self-organizing processing elements can be trained to discover the knowledge embedded in data. A knowledge discovery system has to be able to deal with domain complexity and data noise. In meeting these needs, the neural network approach seems to hold the promise of providing the ultimate solution for knowledge discovery. However, delivering this promise depends on how neural network knowledge is understood in a human sense. Two critical issues are Knowledge representation and knowledge extraction. Future research will seek a better general theory for knowledge extraction and a way to develop a special neural network whose knowledge can be decoded faithfully.