SMS spam collection
5,574 English text messages tagged as spam or legitimate.
Datasets
Wine quality
Physicochemical tests of red and white Vinho Verde wines with a quality score from tasters. A favourite for regression exercises.
Online retail transactions
All transactions of a UK online gift shop over one year: invoice, product, quantity, price and country.
Adult census income
Census records with age, education, occupation and hours worked, labelled by whether income exceeds 50K a year.
Beijing PM2.5
Hourly PM2.5 at the US Embassy in Beijing with weather data, 2010 to 2014. Some PM2.5 values are missing.
Student performance
Grades of secondary school students in Portugal with family, social and school attributes.
Bank marketing campaigns
Phone campaigns of a Portuguese bank: client profile, contact history and whether the client subscribed a term deposit.
Air quality of an Italian city
Hourly responses of a gas sensor array in a polluted area, next to reference CO, NOx and benzene readings.
Seoul bike sharing demand
Number of bikes rented each hour in Seoul, with temperature, humidity, wind and holidays.
Bike sharing, Washington D.C.
Hourly rentals of the Capital Bikeshare system over two years, with weather and calendar information.
Breast cancer diagnosis (Wisconsin)
Features computed from digitised images of breast mass samples, labelled benign or malignant.
Heart disease (Cleveland)
Clinical measurements of 303 patients, such as age, cholesterol, blood pressure and chest pain type, with the diagnosis of heart disease.