[sc34wg3] Unsupervised Methods for Determining Object and Relation Synonyms on the Web

[Patrick Durusau]

Greetings!

The paper in the subject line can be found at:

http://www.cis.temple.edu/~yates/papers/resolver-jair09.pdf

Significant for several reasons:

1) Something to add to the toolbox for seeding topic maps.

2) Interesting work on the "properties" that serve to distinguish subjects.

3) Sets a benchmark that topic maps can better.

4) Its algorithms and processing choices

I don't see this as competing with topic maps as if you are a the person 
most concerned with:

a. a medical diagnosis

b. repairs to a commercial jet

c. correct operation of a nuclear power facility

d. (your use case goes here)

then 78% precision and 68% recall aren't really all that attractive. ;-)

But it is a good starting place and hopefully this paper and related 
research will start to show up in topic map work.

Oh, the formal citation: Journal of Artificial Intelligence Research 34 
(2009) 255-296.

Hope everyone is looking forward to a great weekend!

Patrick

-- 
Patrick Durusau
patrick at durusau.net
Chair, V1 - US TAG to JTC 1/SC 34
Convener, JTC 1/SC 34/WG 3 (Topic Maps)
Editor, OpenDocument Format TC (OASIS), Project Editor ISO/IEC 26300
Co-Editor, ISO/IEC 13250-1, 13250-5 (Topic Maps)