Mallzee Dataset

Binary responses (swipes) to being presented a single item (piece of clothing) in the Mallzee retail app ( The anonymised dataset is comprised of the real responses of 200 users to 500 items (different for each user) as they were interacting with the Mallzee app.

Items are specified by multivariate descriptions; specifically (with dataset labels in parentheses): - the price during the user interaction (CurrentPrice) - the discount from the original price in absolute value (DiscountFromOriginal) - monetary currency in which the price is displayed (Currency) - the type of clothing (TypeOfClothing) - the intended gender for the item (Gender) - whether the item was in stock at the time of interaction (InStock) - the brand of the item (Brand) - the colour of the item (Colour)

"CurrentPrice" and "DiscountFromOriginal" are continuous variables. The value of -1 for "CurrentPrice" represents the lack of a set price during the interaction (rather than a missing value). "InStock" is binary with 2 indicating that the item is in stock, and 1 that it is absent. All other variables are categorical, with each category labelled by an integer, starting from "1".

Users have been anonymised and are represented by a numerical id.

User responses are represented by a binary variable, where 0 indicates rejection (swiping left), and 1 indicates acceptance (swiping right). Accepting an item saves it to a list of items for potential future purchase.

Data and Resources

Additional Info

Field Value
Author Pavlos Andreadis
Maintainer Pavlos Andreadis
Version 1.0
Last Updated January 5, 2022, 14:20 (UTC)
Created November 14, 2016, 17:30 (UTC)
comments powered by Disqus
comments powered by Disqus