R studio
Want help to write your Essay or Assignments? Click here
R studio
Order Instructions:
The data for this week’s assignment is the Breast Cancer data set found in the mlbench library for computing in the R studio
It consists of 699 observations on 11 variables. The names of the variables are (check the help information for the data for some other information).
Id, the code number (not used in the analysis)
cl.thikness, the clump thickness
cell.size, cell size
cell.shape, cell shape
Marg.adhesion, marginal adhesion
Epith.c.size, single epithelial cell size
Want help to write your Essay or Assignments? Click here
Bare.nuclei, bare nuclei
Bl.cromatin, Bland cromatin
Normal.nucleoli, normal nucleoli
Mitoses
Class (this is the target variable)
The goal is to predict values of Class.
Variables 2 through 10 are coded as factors: for some of the work you may want to convert them to numerical values and standardize them.
Want help to write your Essay or Assignments? Click here
In your analysis
Examine the data for missing values, and explain how you decided to deal with them
Examine the distributions of variables 2 through 10 (boxplots or histograms, or both) for shape and outliers. These variables are recorded so that all are scaled to a range of 0 to 10, but their distributions may not be (probably are not) the same, hence the need for investigation. If you want to create histograms, note that the hist() function in R requires a vector of values, so you will need to do something like hist(as.numeric(BreastCancer$cell.size)) to get the histogram to work
Prepare the data for use in a neural network
Fit a single hidden layer neural network to the data and assess its performance. (Do the usual training/testing process). Try to tune the model to select an ‘optimal’ value of hidden nodes
Use network averaging to model the data, and assess its performance
We can write this or a similar paper for you! Simply fill the order form!
