Predicting Pinterest

Automating a Distributed Human Computation


Abstract

Everyday, millions of users save content items for future use on sites like Pinterest, by "pinning" them onto carefully categorised personal pinboards, thereby creating personal taxonomies of the Web. This paper seeks to understand Pinterest as a distributed human computation that categorises images from around the Web. We show that despite being categorised onto personal pinboards by individual actions, there is a generally a global agreement in implicitly assigning images into a coarse-grained global taxonomy of 32 categories, and furthermore, users tend to specialise in a handful of categories. By exploiting these characteristics, and augmenting with image-related features drawn from a state-of-the-art deep convolutional neural network, we develop a cascade of predictors that together automate a large fraction of Pinterest actions. Our end-to-end model is able to both predict whether a user will repin an image onto her own pinboard, and also which pinboard she might choose, with an accuracy of 0.69 (Accuracy@5 of 0.75).


Paper

Changtao Zhong, Dmytro Karamshuk, Nishanth Sastry. Predicting Pinterest: Automating a Distributed Human Computation. Proceeding 24th International World Wide Web Conference (WWW), 2014. [PDF] [Talk slides (PDF)]


Dataset

The Pinterest dataset used in this paper is available for the research community now. Please find out more from here.