Artificial Intelligence Problem

"Artificial Intelligence” is an area of computer science that aims to develop important concepts from simulating human intelligence in order to build models that mimics intelligent behavior. Artificial intelligence is a set of algorithms that are used to extract features and/or find patterns within some collectable data ranging from measurable time series to descriptive statements.
Several expert systems have been developed in various medical fields including medical diagnosis and medical survival rate prediction where previous experience or expert knowledge is often used to design expert systems. The task becomes interesting when no or very little prior knowledge is available or when previous experience is actually provided by data mining algorithms.
The need for an intelligent algorithm for decision making comes from the well-known limits of human information processing power. It has been noticed that the need for support for human decision-makers is due to the

First there is the difficulty of selecting a proper network architecture which is suitable for the task. A good match between network complexity and the complexity of the function to be approximated is required to obtain a good level of generalization [1]. This problem is also known as the model selection problem. The ability to generalize well is the modest important feature of any pattern recognition algorithm.
A second difficulty occurs with the training of FNNs, which is an optimization task. The majority of FNNs employ non-linear hidden neurons because this allows for the creation of more complex models. The use of non-linear neurons results in a non-linear optimization task. Non-linear optimization may often be difficult to perform because of complex error surfaces with many local minima. This may result in large training times, and for some training algorithms the chance of convergence to a poor