The Car Evaluation Dataset originates from a simple hierarchical decision-making model detailed in Expert System for Decision Making, Sistemica 1(1), pp. 145-157, 1990. The model assesses car acceptability based on the following conceptual structure:

CAR (Car Acceptability Based on Price)

• Buying: The purchase price of the car.
• Maint: The cost of maintenance.

TECH & COMFORT (Technical Features and Comfort)

• Doors: The number of doors.
• Persons: Passenger capacity (number of people the car can carry).
• Lug_Boot: The size of the luggage compartment.
• Safety: The estimated safety level of the car.

The dataset directly relates car acceptability (CAR) to six input attributes—buying, maint, doors, persons, lug_boot, and safety—with the structural information removed. This makes the dataset particularly suitable for testing methods related to constructive induction and structure discovery.

Original Version of Data : https://archive.ics.uci.edu/dataset/19/car+evaluation

Cleaned Version in CSV File : https://www.kaggle.com/datasets/rohit265/car-evaluation-uci-machine-learning

Attribute Variables

1. Buying Price (buying):
   • Categories: vhigh, high, med, low.
2. Maintenance Cost (maint):
   • Categories: vhigh, high, med, low.
3. Number of Doors (doors):
   • Categories: 2, 3, 4, 5more.
4. Passenger Capacity (persons):
   • Categories: 2, 4, more.
5. Luggage Boot Size (lug_boot):
   • Categories: small, med, big.
6. Safety Rating (safety):
   • Categories: low, med, high.

Class Labels (Car Acceptability)

• Unacceptable (unacc)
• Acceptable (acc)
• Good (good)
• Very Good (vgood)

This dataset is a valuable resource for evaluating decision-making models and exploring inductive reasoning techniques.

DAX for Power BI Tutorial

Buying_Value = 
SWITCH(
    TRUE(),
    'car'[buying] = "low", 1,
    'car'[buying] = "med", 2,
    'car'[buying] = "high", 3,
    'car'[buying] = "vhigh", 4,
    BLANK()  -- Default case if no match
)