Transcript MYRAW data
DSCI 4520/5240 DATA MINING MYRAW Nonprofit donor data MYRAW Overview Determine who is likely to donate to a non-profit organization campaign and target them for donation solicitation Lecture 2 - 1 The Scenario A nonprofit organization wants to send greeting cards to lapsed donors to encourage them to make a new donation. The organization wants to build a model to predict those most likely to donate. The types of information available include personal information such as age, gender, and income past donation information such as average gift and time since first donation census tract information based on the donor’s address such as the percent of households with at least one member that works for the federal, state, or local government. 2 DSCI 4520/5240 MYRAW data DATA MINING Lecture 2 - 3 DSCI 4520/5240 MYRAW data DATA MINING Lecture 2 - 4 DSCI 4520/5240 Variable roles DATA MINING The first several variables (Age through FirstTime) have the measurement level interval because they are numeric in the data set and have more than 20 distinct levels. The model role for all interval variables is set to input by default. The variables Gender and Homeowner have the measurement level binary because they have only two different nonmissing levels. The model role for all binary variables is set to input by default. Lecture 2 - 5 DSCI 4520/5240 Variable roles DATA MINING The variable IDCode is listed as a nominal variable because it is a character variable with more than two nonmissing levels. Furthermore, because it is nominal and the number of distinct values is greater than 20, the IDCode variable has the model role ID. If IDCode had been stored as a number, it would have been assigned an interval measurement level and an input model role. The variable Income is listed as a nominal variable because it is a numeric variable with more than two but no more than ten distinct levels. All nominal variables are set to have the input model role. You will change the level to ordinal. Lecture 2 - 6 DSCI 4520/5240 Variable roles DATA MINING These variables do have useful information, however, and it is the way in which they are coded that makes them seem useless. Both variables contain the value Y for a person if the person has that condition (pet owner for Pets, computer owner for PCOwner) and a missing value otherwise. Decision trees handle missing values directly, so no data modification needs to be done for fitting a decision tree; however, neural networks and regression models ignore any observation with a missing value, so you will need to recode these variables to get at the desired information. For example, you can recode the missing values as a U, for unknown. You do this later using the Impute node. Lecture 2 - 7