Option #1 Bank Marketing Banks often market their financial products through phone calls. Assume that you are a data mining expert who works in the marketing campaign team at a bank. You are tasked...

Abstract: Banks often come up with new financial products via campaigns through phone calls. The paper is written in such a way that all the descriptive and graphical analysis is done based on individuals’ preferences given in the data. By analysing the data further, we modelled a way if a new product is launched in the market, how will customers make choices. Using the SAS studio, we have built a model to predict the success/failure rate of the launch.
There were total 21 variables, with 11 being the categorical variables and rest being numerical.
1) Summary and graphical statistics:
The frequency of all the categorical variables are as follows:
We can see from above; the maximum number of contacts are from Administrative section.
The martial status of the contacts holds more weightage for Married individuals.
Here, maximum of the individuals holds University degree.
In terms of loan, only single individual holds a credit default, more than 50% holds housing loan and only 16% holds personal loan.
We can see that in the month of May maximum contacts were made. However, it does not matter which specific day of the month an individual was contacted as all the days from Monday to Friday hold almost same weightage.
There exists only 3% success rate in the previous campaign.
Around 65% of the individuals were contacted through cellular and remaining via telephone.
Based on the above graph, there were 3 different outcomes from the previous campaign, i.e., failure, non-existent and success. However, using the bars graph, we show number of contacts made in this campaign in contrast with the previous campaign result. Also, we can see maximum of them were made to Non-Existent; aiming to involve the individuals in this campaign. Pdays refers to the number of days passed by when previous contact was made, which is maximum for the Non-Existent (i.e. sum of more 999 (no previous contact was made), which here means there all the individuals...