Traffic accident database consists of all accidents that happened in Slovenia’s capital city Ljubljana between the years 1995 and 2005.
Adventure Works 2014 (OLTP version) is a sample database for Microsoft SQL Server, which has replaced Northwind and Pub sample databases that were shipped earlier. The database is about a fictious, multinational bicycle manufacturer called Adventure Works Cycles.
Airline on-time data are reported each month to the U.S. Department of Transportation (DOT), Bureau of Transportation Statistics (BTS) by the 16 U.S. air carriers that have at least 1 percent of total domestic scheduled-service passenger revenues, plus two other carrie…
The study STULONG is a longitudinal 20 years lasting primary preventive study of middle-aged men. The study aims to identify prevalence of atherosclerosis RFs in a population generally considered to be the most endangered by possible atherosclerosis com…
The task is to predict rank of teams.
The task is to predict whether the team plays playoff, or not.
This is an older data set of chemical structures containing 328 compounds labeled by their half-life for aerobic aqueous biodegradation (a regression task).
Evaluation of patients on liver disorder.
For prediction of whether a given molecule is carcinogenic or not. The dataset contains 182 positive carcinogenicity tests and 148 negative tests.
Transactional data from Czech debit card company specialising on payments at petrol pumps.
A database containing geospatial information, as well as SAT average scores and Free-or-Reduced-Price Meal eligibility data, for California schools.
The goal is to predict the outcome of a match.
The CiteSeer dataset consists of 3312 scientific publications classified into one of six classes. The citation network consists of 4732 links. Each publication in the dataset is described by a 0/1-valued word vector indicating the absence/presence of the corresponding …
The schema is for Classic Models, a retailer of scale models of classic cars. The database contains typical business data such as customers, orders, order line items, products and so on.
The Consumer Expenditure Survey (CE) collects data on expenditures, income, and demographics in the United States. The public-use microdata (PUMD) files provide this information for individual respondents without any information that could identify respondents. PUMD fi…
The Cora dataset consists of 2708 scientific publications classified into one of seven classes. The citation network consists of 5429 links. Each publication in the dataset is described by a 0/1-valued word vector indicating the absence/presence of the corresponding wo…
The task is to predict "Forest area (% of land area)" for 247 countries in 2012 based on the previous values.
Craft beers labeled by styles and composition. A separate dataset lists breweries by state.
A bit more complex artificial database with loops.
Artificial data from a Czech bank.
Officer-involved shootings as disclosed by the Dallas Police Department. Includes separate tables for officer and subject/suspect information.
The set of positive examples consists of all sentences of up to seven words that can be generated by the DCG in Bratko's book (565 positive examples).The set of negative examples was generated by randomly selecting one word in each positive example and replacing it by …
Dunur is a relatedness of two people due to marriage such that A is dunur of B if a child of A is married to a child of B.
Elti is a relatedness of two people due to marriage such that A is elti of B if A's husband is a brother of B's husband.
The employees test database: small, fake database of employees.
Ergast.com is a webservice that provides a database of Formula 1 races, starting from the 1950 season until today. The dataset includes information such as the time taken in each lap, the time taken for pit stops, the performance in the qualifying rounds etc. of all Fo…
PKDD'99 Financial dataset contains 606 successful and 76 not successful loans along with their information and transactions. The standard task is to predict the loan outcome for finished loans (A vs B in loan.status) at the time of the loan start (defined by loan.dat…
Anonymised data from a hospital in Hradec Kralove, Czech Republic, about treatment and medication.
PAKDD'15 Data Mining Competition: The task is to reconstruct the information about user’s gender from product viewing logs. The data were obtained from simulations of product viewing activities of users with known gender. The data closely follow the real-life distribut…
Data on deputies and senators in the Czech Republic.
KDD Cup 2001 prediction of gene/protein function and localization.
GO Sales dataset from IBM contains information about daily sales, methods, retailers, and products of a fictitious outdoor equipment retail chain “Great Outdoors” (GO). The task is to predict sale quantity.
This dataset includes funding grants from the National Science Foundation. The task is to predict the award amount.
PKDD'02 Hepatitis dataset describes 206 instances of Hepatitis B (contrasting them against 484 cases of Hepatitis C).
The Hockey Database follows the same general design as the Lahman Baseball Database. In addition to the NHL, the Hockey DB covers the following early and alternative leagues: NHA, PCHA, WCHL and WHA. It contains individual and team statistics from 1909-10 through the 2…
The IMDb database: moderately large, real database of movies.
MovieLens data set from the UC Irvine machine learning repository.
The task is to identify, whether the position of two kings and a rook on a chessboard is legal or illegal.
Lahman’s baseball database contains complete batting and pitching statistics from 1871 to 2014, plus fielding statistics, standings, team stats, managerial records, post-season data, and more.
Bulgarian court decision metadata.
PKDD'99 Medical dataset describes 41 patients with Thrombosis.
This domain is about finite element methods in engineering. The task is to predict how many elements should be used to model each edge of a structure. The target predicate is mesh(Edge,Number) where the Number of elements in the Mesh model can vary between 1 and 17.
A geography dataset from University of Göttingen describes 114 Christian countries and 71 non-Christian countries.
The dataset describes a family composed of 86 people across 5 generations. The family dataset includes 744 positive instances and 1488 randomly generated negative instances.
The Musk database describes molecules occurring in different conformations. Each molecule is either musk or non-musk and one of the conformations determines this property. Such a problem is known as a multiple-instance problem, and is modeled by two tables molecule and…
The dataset comprises of 230 molecules trialed for mutagenicity on Salmonella typhimurium. A subset of 188 molecules is learnable using linear regression. This subset was later termed the ”regression friendly” dataset. The remaining subset of 42 molecules is named the …
A sample database from Alchemy website.
A database with information about basketball matches from the National Basketball Association. Lists Players, Teams, and matches with action counts for each player.
2015 NCAA Basketball Tournament.
The Northwind database contains the sales data for a fictitious company called Northwind Traders, which imports and exports specialty foods from around the world.
The National Institute of Diabetes and Digestive and Kidney Diseases conducted a study on 768 adult female Pima Indians living near Phoenix.
A database with information about football matches from the UK Premier League. Lists Players, Teams, and matches with action counts for each player.
A database from The Predictive Toxicology Evaluation Challenge (1997). The task is to predict whether the compound is carcinogenic, or not.
The Pubmed Diabetes dataset consists of 19717 scientific publications from PubMed database pertaining to diabetes classified into one of three classes. The citation network consists of 44338 links. Each publication in the dataset is described by a TF/IDF weighted word …
The pubs sample database is modeled after a book publishing company.
A database of restaurants in San Francisco. The goal is to predict the customer's satisfaction.
The venerable sakila test database: small, fake database of movies.
A simple artificial database in star schema.
The task is to predict whether two given people are from the same generation.
You are a member of the Sales Management team in a large retail bank. The current date is July 02, 2007. Your Sales Director has just asked you to generate additional revenues of $1,500,000 before September 01, 2007. You must find ways to sell more "Credit++" – the ne…
The task is to diagnose power-supply failures in a communications satellite.
Seznam.cz is a web portal and search engine in the Czech Republic. The data represent online advertisement expenditures from Seznam's "wallet". Table description: client: location and domain field of the client (anonymized) dobito: prepaid into a wallet in Czech cur…
The San Francisco Dept. of Public Health’s database of eateries, inspections of those eateries, and violations found during the inspections. The task is to predict the unscheduled inspection scores from 2013 to 2016. The scores range from 1 to 100, where 100 means that…
The Open Source Shakespeare is a collection of Shakespeare's complete works. This is a much more interesting data set than some boring imaginary online retailer. In this dataset, people die! The task is to predict the character, who speaks the lines.
An anonymized dump of all user-contributed content on the Stats Stack Exchange network.
Student Loan contains data about students enrollment and employment status, and the aim is to find rules that define a students' obligation for paying his/her loan back.
Predictive Toxicology Challenge (2000) consists of more than three hundreds of organic molecules marked according to their carcinogenicity on male and female mice and rats.
TPC-C is the benchmark published by the Transaction Processing Performance Council (TPC) for Online Transaction Processing (OLTP).
TPC-D represents a broad range of decision support (DS) applications that require complex, long running queries against large complex data structures.
TPC-DS is the new decision support benchmark that models several generally applicable aspects of a decision support system, including queries and data maintenance. Although the underlying business model of TPC-DS is a retail product supplier, the database schema, data …
TPC-H is the benchmark published by the Transaction Processing Performance Council (TPC) for decision support.
East-West challenge (1980) database describes east-bound and west-bound trains.
An artificial database from Simon Fraser University describing students, professors and courses.
The task is to learn rules that identify the legal states of the U-tube dynamical system.
This dataset lists facts about the Department of Computer Science and Engineering at the University of Washington (UW-CSE), such as entities (e.g., Student, Professor) and their relationships (i.e. AdvisedBy, Publication).
Visual Genome is a dataset, a knowledge base, an ongoing effort to connect structured image concepts to language.
VOC database provides a peephole view into the administrative system of an early multi-national company, the Vereenigde geoctrooieerde Oostindische Compagnie (VOC for short - The (Dutch) East Indian Company) established on March 20, 1602.
Walmart challenges participants to accurately predict the sales of 111 potentially weather-sensitive products (like umbrellas, bread, and milk) around the time of major weather events at 45 of their retail locations.
The WebKB dataset consists of 877 scientific publications classified into one of five classes. The citation network consists of 1608 links. Each publication in the dataset is described by a 0/1-valued word vector indicating the absence/presence of the corresponding wor…
A database of 239 states and their cities.