Use R to solve the following problems from the textbook when appropriate-Compare the results. Which model is preferable?

Assignment Question

Part Use R to solve the following problems from the textbook when appropriate. Refer to RZONE in the textbook for possible hints for R commands. Furthermore, you may need to consult the online documentation of the relevant R packages. Submit a document with your solutions. For the following exercises, use the attached churn data set. For solving the problem you may need to consult the online documentation on the arules package and apriori command and its parameters. Filter out all variables from the dataset except the following: VMail Plan, Intl Plan, CustServ Calls and Churn, while setting CustServ Calls to be ordinal (e.g. using the factor command.) Subsequently, allow the three predictors in either antecedent or consequent (but not both), but do not allow Churn to be in the antecedent. Furthermore, set the minimum support to 1%, the minimum rule confidence to 5%, and the maximum number of antecedents to 1. Use rule confidence as your evaluation measure.

Find the association rule with the greatest lift. Report the following for the rule in (2): Number of instances Support % Confidence % Lift Explain in terms of the rule and data, what each of the measures from (3.b), (3.c) and (3.d) mean.

Part 2 Use R to solve the following problems from the textbook when appropriate. Refer to RZONE in the textbook for possible hints for R commands. Furthermore, you may need to consult the online documentation of the relevant R packages. Submit a document with your solutions. For the following exercises, use the attached churn data set. Apply a CART, C4.5/C5.0 and neural network model for predicting churn data set.4. For each model determine the following measures. Proportion of false positives. Proportion of false negatives. Overall error rate. Overall model accuracy (1 – overall error rate). Sensitivity. Specificity. 5. Compare the results. Which model is preferable?