Exercises

1. MLP has the potential to create prediction models that are more accurate than prediction models that are created by linear regression. This statement is generally correct. In this exercise, we want to explore one of the reasons why the statement is correct. Answer the following questions:

   a) The following formula shows the linear equation that we used to connect the dependent and independent attributes of the number of MSU applications problem. Count and report the number of coefficients that linear regression can play with to fit the equation to the data.

   

   b) Figure 6.8 shows the MLP network structure we used to connect the dependent and independent attributes of the number of MSU applications problem. Count and report the number of connections' weights MLP can play with to fit the network to the data.

   c) Use your answers from a) and b) to state why MLP has more potential in terms of creating prediction models with higher accuracy.

2. In this exercise, we will be...