Jester is a joke recommender system developed at UC Berkeley to study social information filtering.

Version 5.0, launched on April 1, 2015, includes new jokes and algorithms, and a redesigned interface.

Datasets for Research

Datasets containing over 5 million anonymous joke ratings from 150k users are freely available for research use at:

As a courtesy, if you use the data, we would appreciate knowing your name, what research group you are in, and the publications that may result.


Eigentaste: A Constant Time Collaborative Filtering Algorithm [bibtex]
Ken Goldberg, Theresa Roeder, Dhruv Gupta, and Chris Perkins. Information Retrieval, 4(2), 133-151. July 2001.

Eigentaste 5.0: Constant-Time Adaptability in a Recommender System Using Item Clustering [bibtex]
Tavi Nathanson, Ephrat Bitton, and Ken Goldberg. ACM Conference on Recommender Systems (RecSys), Minneapolis, MN, Oct 2007.

Algorithms, Models and Systems for Eigentaste-Based Collaborative Filtering and Visualization [bibtex]
Tavi Nathanson. Master's thesis, EECS Department, University of California, Berkeley, May 2009.

Papers using the Jester Dataset


Jester 5.0 Project Team:
Ken Goldberg, Viraj Mahesh, Sanjay Krishnan, Jay Patel
Jester 4.0 Project Team:
Ken Goldberg, Tavi Nathanson, Ephrat Bitton
Jester 3.0 Project Team:
Ken Goldberg, Robert Hennessy
Jester 2.0 Project Team:
Ken Goldberg, Dhruv Gupta, Chris Perkins
Jester 1.0 Project Team:
Ken Goldberg, Dhruv Gupta, Hiro Narita, Mark DiGiovanni
We also thank Richard Wallace, Hal Varian, Vivek Sanghi, Carol Kirschenbaum, Avi Goldberg, David Pescovitz, Bob Farzin, Adam Jacobs, Derek Poon, and Cathie Walker.


Related Projects


Jay Patel, Viraj Mahesh, Sanjay Krishnan

EECS, UC Berkeley

Ken Goldberg

Professor, IEOR and EECS, UC Berkeley

Mailing List

Join our Mailing List to receive updates when new jokes are added!

Privacy Policy