MidWest Airlines has entered into a partnership with the wireless phone company Telcom to sell the latter's service via direct mail. The MidWest-Airlines data file contains a subset of a data sample...

MidWest Airlines has entered into a partnership with the wireless phone company Telcom to sell the latter's service via direct mail. The MidWest-Airlines data file contains a subset of a data sample who has already received a test offer. About 13% accepted. You are asked to develop a model to classify MidWest customers as to whether they purchase a wireless phone service contract (target variable Phone_sale), that can be used to predict classifications for additional customers.


Sheet1 TopflightBalanceQual_milescc1_miles?cc2_miles?cc3_miles?Bonus_milesBonus_transFlight_miles_12moFlight_trans_12Online_12EmailClub_memberAny_cc_miles_12mo 11205760100588312325020101 12058401003450113450110100 120726010013754000100 118521000012272122720000 1597630100337722010010101 01922100004655850010000 017792601002079713000101 010021000000000000 043832000000000100 112375901005057223250050111 0231930000650455030000 East-West Airlines has entered into a partnership with the wireless phone company Telcon to sell the latter's service via direct mail TERM PROJECT- SCMA 851 PREDICTIVE ANALYTICS (Due on October 8th 2021) Aims: Data analysis, transformation, model development and assessment. Please submit a word document that includes all of your answers for each question and submit your R scripting files with the report. MidWest.csv Prediction Scoring.xlsx MidWest Airlines has entered into a partnership with the wireless phone company Telcom to sell the latter's service via direct mail. The MidWest-Airlines data file contains a subset of a data sample who has already received a test offer. About 13% accepted. You are asked to develop a model to classify MidWest customers as to whether they purchase a wireless phone service contract (target variable Phone_sale), that can be used to predict classifications for additional customers. 1. Explain the role of Data Dictionary below using the CRISP-DM process for a successful predictive analytics project in the MidWest Co. 2. First, transform variables based on the data dictionary given below and form a dataframe in RStudio. Present summary statistics for Balance, Bonus_miles, and Topflight variables. 3. Develop a logistic regression model to predict customer purchase behavior (Target variable= Phone_sale). Report the model output. Which factors are statistically significant predictors of the purchase decision. Present Odds Ratios of the significant variables with 95% Confidence Intervals. Plot the ROC curve of the model. 4. Use a step() function in R for feature selection. Which factors are significant in this final model? Present the ORs for these variables. Plot the ROC curve of the model. 5. Construct a Neural Network model. Change the number of nodes in the hidden layer from 20 to 50 and rerun the model. Plot the ROC curves of both models. 6. Fill the following table for the models developed. MODEL ACCURACY PRECISION SENSITIVITY F1 AUC Logistic regression 1 Logistic regression 2 Neural Network (20 Nodes) Neural Network (100 Nodes) 7. Which model would you suggest marketing department should use for targeting customers? Summarize your results in a discussion. 8. Use one of the models you developed above and the provided records of new customers in the PredictionScoring.xlsx data file for predicting if they would by the Telcom service as a result of the direct mailing campaign or not. Supplement: Data Dictionary FIELD NAME DATA TYPE MAX DATA LENGTH RAW DATA OR TELCOM CREATED FIELD? DESCRIPTION ID# NUMBER Telcom Unique ID Topflight CHAR 1 Raw Indicates whether flyer has attained elite "Topflight" status, 1 = yes, 0 = no Balance NUMBER 8 Raw Number of miles eligible for award travel Qual_miles NUMBER 8 Raw Number of miles counted as qualifying for Topflight status cc1_miles? CHAR 1 Raw Has member earned miles with airline freq. flyer credit card in the past 12 months (1=Yes/0=No)? cc2_miles? CHAR 1 Raw Has member earned miles with Rewards credit card in the past 12 months (1=Yes/0=No)? cc3_miles? CHAR 1 Raw Has member earned miles with Small Business credit card in the past 12 months (1=Yes/0=No)? Bonus_miles NUMBER Raw Number of miles earned from non-flight bonus transactions in the past 12 months Bonus_trans NUMBER Raw Number of non-flight bonus transactions in the past 12 months Flight_miles_12mo NUMBER Raw Number of flight miles in the past 12 months Flight_trans_12 NUMBER Raw Number of flight transactions in the past 12 months Online_12 NUMBER Raw Number of online purchases within the past 12 months Email CHAR 1 Raw E-mail address on file. 1= yes, 0 =no? Club_member NUMBER Telcom Member of the airline's club (paid membership), 1=yes, 0=no Any_cc_miles_12mo NUMBER Telcom Dummy variable indicating whether member added miles on any credit card type within the past 12 months (1='Y', 0='N') Phone_sale NUMBER Telcom Dummy variable indicating whether member purchased Telcom service as a result of the direct mail campaign (1=sale, 0=no sale)