Supermarket Data aggregated by Customer and info from shops pivoted to new columns.

Pennacchioli, D., Coscia, M., Rinzivillo, S., Pedreschi, D. and Giannotti, F., Explaining the Product Range Effect in Purchase Data. In BigData, 2013.

Source

Supermarket data

Stats from etsy.com shops: items, followers, admirers, feedback, open date, sales,...

Source

Data from 58,000 shops extracted directly from etsy.com using import.io crawler

Sample dataset of one year hourly basis machine monitor, with the recorded info about failures.

The Sales Jan 2009 file contains some “sanitized” sales transactions during the month of January. Below is a sample of a report built in just a couple of minutes using the Blank Canvas app. These 998 transactions are easily summarized and filtered by transaction date, payment type, country, city, and geography

Source

spatialkey datasets

A dataset of steel plates’ faults, classified into 7 different types. The goal was to train machine learning for automatic pattern recognition.

Source

Steel Plates Faults Data Set at UCI Machine Learning Repository

BBVA Innova challenge Big Data https://www.centrodeinnovacionbbva.com

BBVA contest. Credit card transactions stats in Barcelona and Madrid between Nov-2012 and Apr-2013

Zipcodes: Give a zone identifier and a commercial category, it returns the top postal codes where the clients with the most payments, unique cards and total spent originate.

Predicted field "Incomes" is the total income in a location (Seller Zipcode) and month, grouped by Commercial Category and Buyer Zipcode.

Data from a semi-conductor manufacturing process

Source SECOM Data Set from UCI

People killed by guns in the USA since Newtown (Dec 14th 2012) until Sept-2013

Sources

• How Many People Have Been Killed by Guns Since Newtown? by Slate and @GunDeaths

• Gun laws in the United States by state by Wikipedia

• The Most Armed States by The Daily Beast

The data is related with direct marketing campaigns of a Portuguese banking institution. The marketing campaigns were based on phone calls. Often, more than one contact to the same client was required, in order to access if the product (bank term deposit) would be (or not) subscribed.

The classification goal is to predict if the client will subscribe a term deposit.

Source

Bank Marketing Data Set at UCI Machine Learning Repository