ANL303: Credit card companies want to encourage their customers to make purchases using their credit cards: Fundamentals of Data Mining Course Work, SUSS, Singapore

(a) Identify the data quality issues in the dataset and perform data cleaning. In less than 100 words, briefly describe how the dataset is cleaned and becomes suitable for data mining using IBM SPSS Modeler. Provide necessary screenshot(s).

(b) Using the dataset obtained from part (a), perform data visualisation that can answer the following enquiries:
• Enquiry 1: What are the top three most popular promotions among zero–balance cardholders?
• Enquiry 2: Is it true that the average income for zero–balance cardholders is higher than the average income for non–zero–balance cardholders in both gender groups?

(i) Construct a k–means model that can provide insights into the characteristics of customers likely to have no card balance. Describe how you identify your model as the final best model. For the model that you have chosen, provide the screenshot of the model’s outputs (showing the cluster summary, and the cluster comparison) and describe the profile of each cluster.

(ii) Discuss how you identify the target customers for ABC based on the clustering results.

(d) Based on part (c), ABC has identified the target customer group for their promotional offerings. Only the target group would be sent promotional information. Construct an association rule mining model that can help ABC understand customer preferences among the promotional offerings within the target group.

Pick two (2) rules that you find interesting. Describe them in terms of support, rule support and confidence, and discuss insights that can help ABC generate revenue.

In your answer, provide screenshots that show clearly (i) which fields are selected for the analysis and their roles, (ii) the parameters used in the algorithm, and (iii) the association rules that you picked for discussion.

(e) Assume that ABC has authorised a new life insurance promotion similar to the previous life insurance promotion. This new life insurance promotion is exclusive for new cardholders who have not had a chance to take advantage of a previous promotion. Based on the dataset described in Table 1, assess the suitability of association analysis for identifying the profile of new cardholders likely to accept the new life insurance promotion.

If it is suitable, discuss (i) which field(s) listed in Table 1 should be included as the antecedent and consequent; (ii) if any data preparation task should be done.