My Photo

Your email address:

Powered by FeedBlitz

April 2018

Sun Mon Tue Wed Thu Fri Sat
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30          
Blog powered by Typepad

Become a Fan

« More Death Cheaper in Future | Main | Six Ticks till Midnight: One Plausible Journey from Here to a Total Surveillance Society »

September 25, 2007


Feed You can follow this conversation by subscribing to the comment feed for this post.

Jeff Carr

Thanks for addressing this topic. I'm interested in your opinion on a recent US Air Force SBIR proposal entitled "Consolidating Entity Information from Heterogeneous Text Sources for Multi-INT Fusion". It concerns the difficulty in solving two cross-document coreference resolution problems: (1) cross-document name disambiguation, and (2) alias resolution. The authors of this topic seem to think that cross-document resolution involving structured and un-structured data across multi-INT domains is still a major problem.

Is that your view as well, Jeff?


Very informative.

Douglas Schwartz

We found the same thing to be true also.


Thank you for this article. It has expanded my narrow thoughts on the uses of Merge and Purge. I had not considered a real-time application of the service.

Umang Juthani

Thank you for this article. My team is currently debating between what kind of business framework we give to our tool set. i.e Suvivorship vs. Order of precedence.
Our goal is to make available individual source system data as well as a golden record computed from various source system holding the most accurate active information about a customer hence a true 360 view of a customer.

Please! expect future question(s) on this topic once i demo this article to my team.


This article makes a strong case for probabilistic databases, or other kind of uncertainty management, and collective matching (current trend in machine learning and data mining). I certainly agree with this view.

The comments to this entry are closed.