Java Data Mining Concepts - Java Data Mining: Strategy, Standard, and Practice - page 131

Table 7-1 Customers Table physical attribute details

Attribute name Data type Attribute description

CUST_ID INTEGER Unique customer identifier

NAME STRING Name of the customer

ADDRESS STRING Address of the customer

CITY STRING City of residence

COUNTY STRING County

STATE STRING State

EDU STRING Educational level, e.g., diploma, bachelor's, master's,

Ph.D.

MAR_STATUS STRING Marital status, e.g., married, single, widowed, divorced

OCCUPATION STRING Occupation of the customer, e.g., clerical, manager,

sales, etc.

INCOME DOUBLE Annual income in thousands of dollars

ETHNIC_GROUP STRING Ethnic group

AGE DOUBLE Age

CAP_GAIN DOUBLE Current capital gains or losses

SAV_BALANCE DOUBLE Average monthly savings balance

CHECK_BALANCE DOUBLE Average monthly checking balance

RETIRE_BALANCE DOUBLE Current retirement account balance

MORTGAGE_AMOUNT DOUBLE Current mortgage/home loan balance

NAT_COUNTRY STRING Native country

CREDIT_RISK STRING Relative credit risk, e.g., high, medium, low

ATTRITE STRING The target attribute indicating whether a customer will

attrite or not. Values include “attriter” and “non-attriter.”

Users may also specify logical attribute characteristics specific to
data mining. For example, physical attribute names in the table or
file can be cryptic, such as, HHSIZE means household size representing
the number of people living as one family. Users can map physical
names to logical names to be more descriptive and hence easier to
understand. Logical data characteristics also include the specification