You might try contacting the Human Rights Data Analysis Group. (hrdag.org) They use semi-supervised machine learning to find duplicate records among tens to hundreds of thousands of records — so, effectively, a classification task with thousands of categories. 
__________________________
Amelia Hoover Green
Assistant Professor
Department of History and Politics
Drexel University

[log in to unmask]
[alt: [log in to unmask]]
http://www.ameliahoovergreen.com




On Jun 28, 2014, at 5:30 PM, Peter Swartz <[log in to unmask]> wrote:

> Dear all,
> 
> I'm interested in a classification problem / supervised-machine-learning
> problem where there are 1000s of categories.  Does anyone know of good
> papers / resources on this subject?
> 
> Thank you!
> Best regards,
> Peter Goodings Swartz
> Data Scientist
> Panjiva
> 
> **********************************************************
>             Political Methodology E-Mail List
>   Editors: Ethan Porter        <[log in to unmask]>
>            Gregory Whitfield   <[log in to unmask]>
> **********************************************************
>        Send messages to [log in to unmask]
>  To join the list, cancel your subscription, or modify
>           your subscription settings visit:
> 
>          http://polmeth.wustl.edu/polmeth.php
> 
> **********************************************************


**********************************************************
             Political Methodology E-Mail List
   Editors: Ethan Porter        <[log in to unmask]>
            Gregory Whitfield   <[log in to unmask]>
**********************************************************
        Send messages to [log in to unmask]
  To join the list, cancel your subscription, or modify
           your subscription settings visit:

          http://polmeth.wustl.edu/polmeth.php

**********************************************************