ABSTRACT
There are many prediction models that have been used, such as the Markov model, decision trees, artificial neural networks (ANNs), support vector machines (SVMs), association rule mining (ARM), and many others. Each of these models has its strengths and weaknesses. However, there is a common weakness among all of these techniques, which is the inability to suit all applications. The reason that there is no such ideal or perfect classifier is that each of these techniques is initially designed to solve specific problems under certain assumptions.
