Home

Data Overview

What you'll see here

Profile the AllLife Bank customer dataset, classify every column, and flag any anomalies before modeling.

Total customers
0
13 predictors
Accepted loan (1)
0
0.00% conversion
Did not accept (0)
0
100.00%
Class imbalance
0.0%
Use stratified split + Recall/F1
Missing values
0
None — no imputation needed
Duplicate rows
0
By customer ID
Negative Experience
0
Anomalous — must be treated
Features used in model
11
ID + ZIPCode dropped
Identifier
ID

Unique customer ID — dropped before modeling.

Demographic
Age

Years.

Experience

Years of professional experience (some negative — anomalous).

Family

Family size (1–4).

Education

1 = Undergrad, 2 = Graduate, 3 = Advanced/Professional.

ZIPCode

Geographic — treated as categorical, dropped for baseline modeling.

Financial
Income

Annual income ($K).

CCAvg

Average monthly credit-card spend ($K).

Mortgage

Mortgage value, $K (0 = no mortgage).

Bank relationship
Securities_Account

Binary 0/1.

CD_Account

Binary 0/1.

Online

Binary 0/1.

CreditCard

Binary 0/1.

Target
Personal_Loan

Binary target — 1 = accepted.

Modeling decisions baked into the data layer

ZIPCode is geographic and high-cardinality — treated as categorical, dropped for the baseline tree to avoid overfitting on noise.

Binary 0/1 columns (Securities, CD, Online, CreditCard) are not one-hot encoded — they are already model-ready.

Negative Experience values are flagged for treatment, not silently ignored.

Class imbalance (0.0% positives) means the train/test split is stratified on Personal_Loan.

AI: Data readiness assessment

Click Generate to produce an evidence-based observation, insight, recommendation, and business benefit for this section.