Artificial neural networks (ANNs) are biologically inspired. Specifically, we borrow ideas from the way the human brain works. The human brain is made up of special cells called neurons. Estimates of the number of neurons in the human brain cover a wide range (up to 150 billion), and there are more than 100 types of neurons divided into groups called networks. Each network contains thousands of highly interconnected neurons. Therefore, the brain can be thought of as a collection of neural networks.Today's ANN, whose application is called neural computing, uses a very limited set of concepts from biological neural systems. The goal is to simulate a large parallel process with interconnected processing elements in a network architecture. Artificial neurons receive inputs similar to the electrochemical impulses that biological neurons receive from other neurons. The output of an artificial neuron corresponds to the signal sent by the biological neuron. These artificial signals can be modified like signals from the human brain. Neurons in ANN receive information from other neurons or external sources, transform or process that information, and pass it to other neurons or as an external output. The way ANN processes information depends on its structure and the algorithms used to process the information. Benefits and applications of neural networks The value of neural network technology includes its usefulness for pattern recognition, learning, and interpretation of incomplete and "noisy" inputs. Neural networks may provide some of the human characteristics of problem solving that are difficult to simulate using DSS or expert system logical analysis techniques. One of these features is pattern recognition. Neural networks can analyze large amounts of data to establish patterns and characteristics in situations where logic and rules are unknown. An example is applying for a loan. By reviewing the many past cases of applicant questionnaires and the "yes or no" decisions made, ANN can create "patterns" or "profiles" of applications that should be approved or rejected. You can then match the new application with the pattern on your computer. When it gets close enough, the computer classifies it as "yes" or "no." If not, it goes to humans for a decision. Neural networks are especially useful for financial applications such as determining when to buy or sell stocks, predicting bankruptcy, and predicting exchange rates. Beyond its role as an alternative computing mechanism, data mining can combine neural computing with other computer-based information systems to create powerful hybrid systems. Neural computing has emerged as an effective technology in pattern recognition. This feature has been transformed into many applications and may even be integrated with fuzzy logic.