The Car Evaluation Dataset from the UCI Machine Learning Repository contains 1,728 samples and 6 categorical attributes. It evaluates car acceptability based on factors such as price, maintenance, and safety, and is an introductory dataset for multiclass learning.
A structured decision model dataset, suitable for learning multiclass classification and decision theory
The data is generated based on a hierarchical decision model, covering evaluation dimensions such as purchase price, maintenance cost, comfort, and safety.
All 6 input features are ordinal categorical variables (low/med/high/vhigh), suitable for learning ordinal encoding.
Cars are rated across four levels: unacc (unacceptable), acc (acceptable), good (good), and vgood (very good).
70% belong to the unacc class, while vgood accounts for only 3.8%, making it excellent material for learning imbalanced classification problems.
The hierarchical decision structure makes it a perfect dataset for learning decision trees and rule extraction.
From the UCI Machine Learning Repository, designed by Marko Bohanec based on decision theory.
From basic classification to decision theory research, with diverse application scenarios
The hierarchical attribute structure is highly suitable for learning decision tree and rule extraction algorithms
A four-class evaluation task for practicing multiclass algorithms such as SVM, KNN, and Naive Bayes
Clear class imbalance for practicing methods such as SMOTE and weighted loss
Ordinal categorical features are suitable for practicing ordinal encoding and feature engineering techniques
Below are sample rows from the car evaluation dataset
buying,maint,doors,persons,lug_boot,safety,class vhigh,vhigh,2,2,small,low,unacc vhigh,vhigh,2,2,small,med,unacc vhigh,vhigh,2,2,small,high,unacc vhigh,vhigh,2,2,med,low,unacc vhigh,vhigh,2,2,med,med,unacc vhigh,vhigh,2,2,med,high,unacc vhigh,vhigh,2,2,big,low,unacc vhigh,vhigh,2,2,big,med,unacc vhigh,vhigh,2,2,big,high,unacc
From browsing to analysis, start your data science project in just minutes
View dataset details on the Ace Data Cloud platform to learn about metadata such as field descriptions, sample size, and license agreement.
After purchase, download the ZIP delivery package, which includes the original CSV (52 KB), source attribution, and CC BY 4.0 license text; field names are added to the official headerless CSV data.
Use pandas.read_csv() to load the data, together with OrdinalEncoder to encode ordinal categorical features.
A classic multiclass dataset with previewable license information and data samples, available for download after purchase. Its hierarchical decision structure makes it an ideal introductory dataset for decision tree learning.
