Have deviating rates or can be freed from the monthly fee due to VIP status or otherparticularities.Ī customer can have any number of current accounts, including zero, and any mix oftypes.Īccounts have an opening and an ending date. Who Wants My Product? Affinity-Based Marketing 79 Basically, each type of account comes with certain fixedmonthly fees and interest rates for credit and debit amounts, but some customers can CH04 is the new type that is to be pushedby our marketing action. The bank now has four types of current (checking) accounts on offer, differentiated bytheir internal names CH01 through CH04. Thus we conduct a number of interviews with our banks business experts and learn thefollowing: The purpose of the CRISP-DM Business Understanding phase is to thoroughly under-stand the task at hand from the perspective of the business users: what terminology dothey use, what do they want to achieve, and how can we assess after the project whetherits goals have been met?
RAPIDMINER STUDIO TUTORIAL HOW TO
Finally, in the lastpart of this chapter, we will discuss how to apply the same methodology to other businesstasks. We can then serve the marketing departmentby delivering the top 20 percent of non-buyers from our final ranking. We will alsosee how to decide which model is most useful. Thus, in what follows, a number of mining models will be developed that each deliver aranking of non-buyers in which the top-ranked customers are those for which the model ismost confident that they ought, in fact, to be buyers (if only they knew it!). Fortunately, most algorithms can do that by delivering a rank-ing of customers, with higher-ranked customers being predicted to be buyers with higherconfidence or probability than lower-ranked ones. Therefore, it is crucial that our data mining algorithm beable to provide that scale. Trying to keep buyers and non-buyers apartin order to find their similarities may sound paradoxical however, difference and simi-larity belong to the same scale.
Assuming that we have gooddata, we can use a standard data mining method, namely binary classification, to try todifferentiate between buyers and non-buyers. Section 7.4 discusses what propertiesmight be useful, and how to build data that reflect them. Our main challenge, therefore, is to identify customer properties that can help us to findsimilarities and that are available in the banks data. Our hope is that the more similar theyare, the higher their affinity. Therefore, we search for customers who have not yet bought it (the non-buyers)but who are similar to the buyers in other respects. We assume those customers who have alreadybought the product (the buyers) to be representative of those who have a high affinity towardthe product.
Thus, how can we determine whether a customer has a high affinity for our new product?We can only use an indirect way of reasoning. We will walk step by stepthrough the fictitious sample data, which is based on real data structures in a standarddata warehouse design, and the RapidMiner processes provided with this chapter to explainour solution. We will follow a simple methodology: the Cross-Industry Standard Process for datamining (CRISP-DM), whose six stages will be applied to our task in the followingsubsections in their natural order (although you would typically switch back and forthbetween the stages when developing your own application). This chapter will explain how to address the business task sketched above using datamining. However, in order not to waste effortson customers who are unlikely to buy, they would like to address only those 20 percent ofcustomers with the highest affinity for the new product. Mail to customers who have not yet opted for it. Sometime after the product is released to the market, a numberof customers have opened accounts of the new type, but many others have not yet done so.The banks marketing department wants to push sales of the new account by sending directħ8 RapidMiner: Data Mining Use Cases and Business Analytics Applications 96ĬRISP-DM - Cross-Industry Standard Process for Data MiningDWH - Data WarehouseBI - Business IntelligenceĪ bank introduces a new financial product: a type of current (checking) account withcertain fees and interest rates that differ from those in other types of current accountsoffered by the same bank. 897.5.4 Confidence Values, ROC, and Lift Charts. 877.5.1 Continuous Evaluation and Cross Validation. Viadee IT-Consultancy, Munster/Koln (Cologne), GermanyĪcronyms. Who Wants My Product? Affinity-BasedMarketing
0 kommentar(er)
