California Housing Dataset:
Regression Analysis Classic Benchmark
A classic regression dataset published by Pace & Barry in 1997, based on the 1990 U.S. Census data. 20,640 neighborhoods, 8 features—covering geographic, economic, and demographic information, making it an ideal choice for learning regression models and housing price prediction.
Dataset Highlights
The California housing price dataset is one of the most popular benchmark datasets in the field of regression analysis.
Regression Benchmark
A standard housing price prediction regression task, suitable for benchmarking various regression algorithms such as linear regression, decision trees, and gradient boosting.
Geographic Information
Includes latitude and longitude coordinates, supports spatial analysis and geographic visualization, exploring the relationship between housing prices and geographic locations.
Economic Features
Median income, housing age, and other socioeconomic indicators reflect the multidimensional influencing factors of the real world.
Moderate Size
20,640 records, suitable for quick experiments and benchmarking, not too small to cause underfitting, and not too large to affect iteration speed.
No Missing Values
The data is complete and clean, ready to use without complex data cleaning and missing value handling steps.
Public and Free
Public domain license, freely usable and redistributable, suitable for learning, teaching, research, and commercial projects.
Applicable Scenarios
From classroom exercises to production models—common uses of the California housing price dataset.
Housing Price Prediction
Build regression models to predict median housing prices, understanding the combined impact of multiple features on housing prices.
Spatial Data Analysis
Utilize latitude and longitude coordinates for geographic visualization, discovering regional distribution patterns of housing prices.
Regression Model Comparison
Compare the MSE and R² performance of models such as linear regression, random forests, and XGBoost on standard data.
Feature Engineering Teaching
An ideal practice dataset for learning feature scaling, interaction features, polynomial features, and other feature engineering techniques.
Data Preview
Sample examples of the California housing price dataset (CSV format)
MedInc,HouseAge,AveRooms,AveBedrms,Population,AveOccup,Latitude,Longitude,MedHouseVal 8.3252,41.0,6.984,1.024,322.0,2.556,37.88,-122.23,4.526 8.3014,21.0,6.238,0.972,2401.0,2.110,37.86,-122.22,3.585 7.2574,52.0,8.288,1.073,496.0,2.802,37.85,-122.24,3.521 5.6431,52.0,5.817,1.073,558.0,2.547,37.85,-122.25,3.413 3.8462,52.0,6.282,1.081,565.0,2.181,37.85,-122.25,3.422
3 Steps to Get Started Quickly
From browsing to using, it only takes a few minutes
Browse the Dataset
View detailed descriptions, field definitions, and data previews of the California housing dataset on the Ace Data Cloud platform.
Download the CSV File
One-click download of a 1.8 MB CSV file to your local machine, no registration, no payment, get it instantly.
Load and Use
Load the data using Python, R, or any data analysis tool, and start training regression models or plotting spatial distribution maps.
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.ensemble import GradientBoostingRegressor
from sklearn.metrics import mean_squared_error, r2_score
# Load data
df = pd.read_csv("california_housing.csv")
# Split features and target variable
X = df.drop("MedHouseVal", axis=1)
y = df["MedHouseVal"]
# Split training and testing sets
X_train, X_test, y_train, y_test = train_test_split(
X, y, test_size=0.2, random_state=42
)
# Train gradient boosting regression model
model = GradientBoostingRegressor(
n_estimators=200, max_depth=5, learning_rate=0.1, random_state=42
)
model.fit(X_train, y_train)
# Predict and evaluate
y_pred = model.predict(X_test)
mse = mean_squared_error(y_test, y_pred)
r2 = r2_score(y_test, y_pred)
print(f"Mean Squared Error (MSE): {mse:.4f}")
print(f"R² Score: {r2:.4f}")
Start Your Regression Analysis Journey
The California housing dataset is one of the most commonly used regression benchmarks in the global data science community. Download for free and start exploring now.