Constraint-Based Association Rule Mining in WEKA

To generate an ARFF file to improve a customer's shopping behaviour and optimize Product placement with the following Attributes.

Age group -> Values - > Teenage, Adult, Senior

Gender -> M, F

Income Level -> low, medium, high

Shopping day -> Weekday, Weekend

Payment Method -> UPI, Cash, Card

Product type -> Electronic items, clothes, groceries, cosmetics

Purchase decision -> yes/no

Generate an associate rule to discover frequent item sets with the following patterns.

1. Identify product categories that are frequently purchased together
2. The Relationship Between income level, shopping day, and payment method.
3. Predict the customer behaviour, try using an association rule, such as if income low, product type = grocery, and which implies purchase = yes

Example

@relation customer_shopping_behaviour

@attribute AgeGroup {Teenage,Adult,Senior}
@attribute Gender {M,F}
@attribute IncomeLevel {Low,Medium,High}
@attribute ShoppingDay {Weekday,Weekend}
@attribute PaymentMethod {UPI,Cash,Card}
@attribute ProductType {Electronic_items,Cloths,Grocery,Cosmetics}
@attribute PurchaseDecision {Yes,No}

@data
Teenage,M,Low,Weekend,UPI,Grocery,Yes
Adult,F,Medium,Weekday,Card,Cloths,Yes
Adult,M,High,Weekday,Cash,Electronic_items,Yes
Senior,F,Low,Weekend,UPI,Grocery,No
Adult,M,Medium,Weekend,Card,Cosmetics,Yes
Teenage,F,Low,Weekday,Cash,Grocery,Yes
Adult,F,High,Weekend,UPI,Electronic_items,Yes
Senior,M,Medium,Weekday,Card,Grocery,Yes
Adult,M,Low,Weekday,UPI,Grocery,Yes
Teenage,F,Medium,Weekend,Card,Cloths,Yes