[Corpora-List] Cross-document coreference/Entity Resolution: $50,000 Spock Challenge

Christopher Manning manning at stanford.edu
Wed Apr 18 05:31:00 CEST 2007


Here's an opportunity for anyone interested to apply their entity
resolution techniques....



$50,000 Spock Challenge

Spock (http://www.spock.com/), a search application for people based
in Silicon Valley, is announcing the launch of The Spock Challenge.
Spock has selected one of its most interesting problems, namely
Entity Resolution, to share with the technical community, allowing
other leading computer scientists and engineers to compete in an open
contest . The winners of this global competition will reap a reward
of $50,000 cash. Spock will also be issuing prizes for 2nd and 3rd
The crux of the Entity Resolution problem is clustering web documents
accurately. For example, how do you programmatically determine
whether a web document about Michael Jackson describes the singer, or
the football player? In order to constrain the problem so that it
can be successfully solved by an individual or a small team, Spock is
providing real world data with ground truth annotations. This data
contains 100,000 documents about people, and the challenge is to
determine all the distinct people described in the data set. This
data can be your training set. Once you’ve got your basic algorithm
working against the training set, Spock will let you further tune
your code by running it against a second test data set.

Participants can compete in the challenge individually or in teams.
The competition will last 4 months and a leader board will be set up
so that participants can submit their program and see how they rank
compared to other submissions.

More details can be found at http://challenge.spock.com . To contact
Spock, e-mail info at corp.spock.com

More information about the Corpora-archive mailing list