12. Refer to the lecture output on neural network model fits…

12. Refer to the lecture output on neural network model fits on the bankruptcy data. https://yanyudm.github.io/Data-Mining-R/lecture/9.A_NeuralNet.html We now fit a glm (model 1), neural network with “hidden=c(3)” (model 2), and neural network with “hidden=c(4)” (model 3) to the bankruptcy data.  We have 10 financial ratios as the predictors (X). The response (Y) is binary, 1=bankruptcy, 0=non-bankruptcy.read.csv(file = “https://yanyudm.github.io/Data-Mining-R/lecture/data/bankruptcy.csv”, header=T) Note that in the output layer, there is only ONE class where DLRSN (deletion reason)=1 is bankruptcy, and DLRSN=0 is nonbankruptcy as a reference class. How many parameters are involved for the neural network model with “hidden=c(3)”?

Question 22:  Use the unscaled data seed1, perform k means c…

Question 22:  Use the unscaled data seed1, perform k means clustering analysis, with k=3, and draw the cluster plot using R function >fviz_cluster() in the R package “factoextra”. Please screenshot your figure here. If your “factoextra” package does not load well, you may use R function >plotcluster() in the R package “fpc” instead.

Question 25:  Use the scaled data seed1scale, perform k mean…

Question 25:  Use the scaled data seed1scale, perform k means clustering analysis, with k=3, and draw the cluster plot using R function >fviz_cluster() in the R package “factoextra”. Please screenshot your figure here. If your “factoextra” package does not load well, you may use R function >plotcluster() in the R package “fpc” instead.