ABSTRACT
Many prediction models 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, they have a common weakness, which is the inability to suit all applications. There is no such ideal or perfect classifier because each of these techniques was initially designed to solve specific problems under certain assumptions.
